Skip to main content

Full text of "The painter's methods & materials : the handling of pigments in oil, tempera, water-colour & in mural painting, the preparation of grounds & canvas, & the prevention of discolouration, together with the theories of light & colour applied to the making of pictures"

See other formats

:-aKissSi:^^^!»sa«^'M: ■ct«aKi['iir>r*'Hgg« 








l l » U«k><' W' 



iU 1U\S COLLECE>r-5^N^£)^ 





A. P. LAURIE, M.A.(Cantab.) 

D.Sc.(Edin.), F.R.S.E., F.C.S., H.R.S.A. 

Projesior of Chemistry to the Royal Academy of Arts, London 

Principal of the Heriot-Watt College, Edinlurgb 


Seeley, Service & Co. Limited 
196 Shaftesbury Avenue 

Uy 53 









This book has been written for the craftsman painter 
in oil, water colour, tempera, and fresco, not for the 
scientific chemist or the manufacturer. It therefore 
deals with methods and the properties of materials 
rather than their chemical description or methods of 
manufacture. For example, the chemistry of the drying 
of linseed oil is so complex as to be unintelligible to 
anyone who is not a student of chemistry. I therefore 
treat only with the results of these changes. 

Those who wish to pursue their studies further may 
consult, among standard books. Sir Arthur Church's 
Chemistry of Paints and Painting and Hurst's Paints, 
Colours, Oils, and Varnishes, and of modern works, 
Varnishes and their Components, by Dr. Morrell, Mal- 
materialien Kunde Als Griindlage Der Maltechnik, Die 
Anorganischen Farbstoffe, and Die Fette Ole, by Professor 

For those who wish to study the materials and 
methods of the Middle Ages to the close of the sixteenth 
century, the Classic work is still Eastlake's Materials 
for the History of Oil Painting, to which may be added 
Hendrie's Translation of Theophilus, Mrs. Merrifield's 
Fresco Painting and Original Treatises on the Arts of 
Painting, Lady Herringham's translation of Cennino 
Cennini, the translation of Vasari on Technique, by Louisa 



Maclehose, and Professor Berger's Beitrdge Zur Ent- 
wickelung' s Geschichte der Maltechnik. May I also be 
permitted to add my Materials of the Painter's Craft 
and the Pigments and Mediums of the Old Masters? 
Students should also consult the Transactions of the 
Tempera Society. 

On sitting down to write this book I found so many 
practical questions arising to which I could discover 
no answer, that I considered it necessary to carry out 
a series of experiments in several different directions, 
the results of which are incorporated in these pages. 
To the best of my knoAvledge the experiments on the 
transparency of pigments in different refractive media 
and the change in the refractive index of linseed oil in 
process of drying are new. 

In numerous other cases the statements in the text 
are the results of actual experiments, though no details 
are given. 

A logical order of treatment of the subject proved 
difficult, and cross-references inevitable. In many cases 
repetition was found necessary to give completeness to 
the particular subject under discussion and save trouble 
to the reader. 

Finally, my thanks are due in the first place to 
Profesor Eibner, not only for his help as a correspondent, 
but also for the assistance obtained from his publica- 
tions, an assistance which occurs so often that I have 
not attempted to acknowledge it in every case in the 
text. My thanks are also due to Mr. Balsillie, of the 
Mineralogical Department of the Scottish National 
Museum ; Mr. Batten ; Mr. Caw, Custodian of the 



Scottish National Gallery ; Mr. Davidson, of the 
Cookson Lead and Antimony Company ; Mr. John 
Duncan, R.SA. ; Mrs. Sargent Florence, for permission 
to reprint her description of "Fresco Painting" from 
the Transactions of the Tempera Society ; and to Dr. 
Morrell : Professor Peddie ; Mr. Charles Sims, R.A. ; 
Mrs. Traquair ; and the Director of the Rijks Museum 
and the Staffs of the Physical and Chemical Departments 
of the Heriot-Watt College. 



I. The Mediums Used in Painting . 

II. The Written Evidence on Early Painting 
Methods in Oil .... 

III. The Written Evidence on Early Painting 

Methods in Oil {contd.) . 

IV. Wood Panels and Paper 
V. Priming of Panels and Canvas 

VI. The Pigments Used in Painting . 

VII. The Pigments Used in Painting {contd.) 

VIII. The Behaviour of White Light . 

IX. Colour and the Prism 

X. Linseed Oil, Walnut Oil, and Poppy Oil 

XL The Optical Properties of Oil 

XII. How TO Paint Oil Pictures 

XIII. Balsams, Resins, Varnishes, Etc. 

XrV. How TO Paint in Tempera . 

XV. Emulsions ...... 

XVI. Fresco Painting ..... 

XVII. Modern Methods of Fresco Painting . 

XVIII. Other Methods of Wall Painting 

XIX. Preservation and Cleaning of Pictures 

XX. Conclusion ...... 

Index ....... 

























p. 13. For W. Sims, R.A. , read Charles Sims, R.A. 

,, For Velasques, read Velazquez, 
pp. 14, 153, 2+g. Text and Illustration. For Tourment, read Fourment. 
pp. 14, 154. Text and Illustration. For Marchesa, read Marchese. 
pp. 14, 162. Illustration. For "Christ knocking at the Door," read "The Light 

of the World." 
pp. 28, 37, 43, 245. For Le Begue, read Le Begue. 
p. 34. For eempera, read Tempera, 
p. 47. For reeipes fos, read recipes for. 
pp. 02, 63. For Gesso-Sottili, read Gesso Sottile. 

p. 72. Illustration. For National Gallerj-, read National Gallerj- of British Art. 
p. 150. Text and Illustration. For Velasquez, read Velazquez. 
pp. 230, 231. Illustrations. For Franz Hals, read Frans Hals. 



1. Professor Hay, By W, Sims, R.A. Frontispiece 


2. Holy Family. By the Master of the " Death of the P^s^ 

Virgin" ....... 40 

3. Christ and the Apostles. By Fra Lippo Lippi . 48 

4. Unfinished Picture. By Sir Davad Wilkie . . 64 

5. Ophelia. By Millais ...... 72 

6. Microphotograph of White Lead ... 80 

7. J. Stuart Traill. By Jolin Lonsdale ... 88 

8. Madam Pompadour. By Frangois Boucher . . 95 

9. Cracks in an Oil Picture . . . . .134 

10. A Woman Bathing. By Rembrandt . . .136 

11. Magnified Photograph of Rembrandt's Brush- 

WORK 137 

12. Lady and Child. By Romney .... 142 

13. Magnified Photograph of Romney's Brush-Work 143 

14. A Section through the Gesso of the Holy Family 144 

15. Santa Barbara. By Jan Van Eyck . . .145 

16. Unfinished Picture. By Cima da ConegUano . 146 

17. Madonna and Child. By Michael Angelo . . 147 

18. The Holy Family. Attributed to Correggio . 148 

19. Portrait of a Lady ...... 149 

20. The Rokeby Venus. By Velasques . . . 150 

21. The Entombment. Bj^ Michael Angelo . . .151 

22. Interior of a Dutch House. By Pieter de Hooch 152 



PLATE Facing page 

23. SusANNE TouRMENT. By Rubens .... 153 

24. Marchesa Giovanni Battista Cattaneo. By 

Van Dyck ....... 154 

25. The Philosopher. By Rembrandt . . .155 

26. Portrait of a Gentleman. By Gerard Terborch . 156 

27. Mr. and Mrs. Lindow. Bj' Romney . . . 158 

28. Magnified Photograph of Romney's Brttsh-Work 159 

29. Portrait of the Artist's Son, Titus. By 

Rembrandt . . . . . . .160 

30. Magnified Photograph of Rembrandt's Brush- 

WORK 161 

31. Christ Knocking at the Door. By Holman Hunt 162 

32. The Age of Innocence. By Reynolds. . . 168 

33. Micro-Photograph of Mastic Varnish . .170 

34. The Rape of Helen. By Benozzo Gozzoli . .184 

35. Venus and Mars. By Botticelli .... 185 

36. Portion of a Fresco. By Giotto . . . 191 

37. BuoN Fresco. By Benozzo Gozzoli . . . 192 

38. BuoN Fresco, By Benozzo Gozzoli . . . 193 

39. Eve. By Michael Angelo 200 

40. Fresco Painting. By Ghirlandaio . . . . 208 

41. A Magnificent Fresco, from the Chapel of the 

RiccARDi. By Benozzo Gozzoli . . . 217 

42. Wall Painting St. Mary's Cathedral, Edinburgh. 

By Mrs. Traquair ...... 224 

43. The Governors of St. Elizabeth's Hospital. By 

Franz Hals 230 

44. The Governors of St. Elizabeth's Hospital 

{partially cleaned) ...... 231 





The mediums used by artists to mix with their pigments 
are very much the same to-day as they have been through 
long periods of time. 

For wall-painting the pigments, merely mixed with 
water, are laid on the wet lime surface of the plaster, 
the binding medium being the crystallized carbonate 
of lime, which is slowly formed by the combination of 
the carbonic acid gas in the air with the lime. Mr. 
Noel Heaton has shown that this method was used in 
painting the frescoes in the Palace of Knossos, and 
the weight of evidence and research is in favour of the 
same method having been used at Pompeii. The frescoes 
of the Italian Renaissance were painted by this method. 
Technical details have varied, but the principle of using 
the carbonate of lime to form the binding material is 
very ancient. 

Gum-arabic was used by the Egyptians, and is used 
by artists to-day for water-colour painting, the pigments 
being ground in this medium. 

Size was used by the Egyptians, and by the Greeks 
and Romans, and throughout the Middle Ages for wall 
decorations. To-day it is used principally by the 
house-painter and the scene-painter. 

The history of the egg medium is more obscure. Its 
B 17 


use for certain special purposes is mentioned by Pliny 
(a.d. 23-79) and in manuscripts of the Middle Ages ; 
but it is not till we come to the treatise on Painting by 
Cennino Cennini, at the beginning of the fifteenth century, 
that we find a full and detailed description of the use of 
the yolk of the egg as a painting medium. The pictures 
of the painters of Italy during the fourteenth and 
fifteenth century, and probably earlier, were principally 
executed in this medium. 

The first mention of the use of the drying vegetable 
oils as media for painting occurs in the manuscripts of 
the eleventh or twelfth century. There is evidence of 
a long northern tradition in the use of this medium 
before the fifteenth century, after which it gradually 
replaced egg and became the universal medium. 

Only one medium used in classical times has fallen 
out of use, namely, beeswax. Pliny tells us how this 
medium was used. The pigments were stirred in with 
the melted wax, and the work executed partly with the 
brush and partly with bronze modelling tools. In order 
to paint with this medium in a cold climate the panel or 
canvas must be artificially warmed. Each stroke of the 
brush must be rapidly put in place and cannot be altered. 

Mr. Burns has painted quite successfully in this 
medium, the canvas being placed with its back to a 
hot fire. The finished picture, when polished with a 
cloth, closely resembles an oil painting. Wax is a fairly 
permanent medium, but readily collects dust and dirt. 
Mr. Burns tells me that his picture painted twenty years 
ago is in excellent condition ; but owing to the wax 
accumulating dirt, he varnished it some time ago, 
quite successfully, with copal oil varnish. This con- 
firms the statement made in the fifth century by the 
physician ^tius, that wax pictures should be varnished 
with a drying oil. 



Examples of wax pictures, on the whole in excellent 
condition, which were found by Professor Flinders 
Petrie in Egypt, are of about the second or third 
century, and are to be seen in the National Gallery. 
With the exception of such pictures preserved by the 
Egyptian sand, I am not aware of any others. Possibly 
the exam.ination now going forward of the Russian 
ikons may reveal some other early examples of wax 
paintings. Encaustic painting with wax dissolved or 
emulsified in a volatile medium was at one time fashion- 
able ; it is quite different from the classical technique. 
An interesting account of the attempts made to revive 
wax painting will be found in the History and Methods 
of Ancient and Modern Painting, by James Ward. 

The Lucca manuscript of the eighth century mentions 
only two mediums for painting — wax for painting on wood 
and glue for painting on parchment. I have already 
stated that the first description of a vegetable drying 
oil as a painting medium occurs in a manuscript of the 
eleventh or twelfth century. This suggests the possibility 
that the use of drying oils for painting was discovered 
some time between the eighth and eleventh century. 

It is sufficient for our purpose that the mediums 
used by artists to-day have the tradition of centuries 
of use behind them, and that modern chemistry has not 
up to the present discovered any other medium, with 
the exception of casein, that has been found suitable 
by painters. 

Of these mediums the most important is oil, and it 
might be supposed that after so many centuries of use 
there would be nothing new to say of the properties of 
this medium or of its correct use. Unfortunately owing 
to loss of studio tradition, less is known to-day of some of 
the properties of this medium and of its correct use than 
was known to the Van Eycks and their followers, and the 



main purpose of this book is to ffiscuss the properties of 
the oil medium and how it should be used. The drying 
oils first known to painters were linseed oil, expressed 
from the seed of the flax, and walnut oil extracted from 
the kernels of the walnut. Hempseed oil, is also men- 
tioned. Later on poppy oil was added to the list. 
To-day walnut oil is little used, linseed oil and poppy 
oil being used in grinding the pigments by the modern 
artists' colourmen. 

Light, the oxygen of the air, and water vapour con- 
vert these oils when exposed in thin layers into a tough, 
elastic, transparent solid which consists principally of 
a substance which the chemists have named linoxin. 
We shall have to discuss this drying process in more 
detail later on, as the conditions of drying must be 
closely studied to avoid the danger of cracking. The 
film of dried oil has two properties which are disturbing 
to the painter. It becomes of a brownish yellow with 
age, and it apparently has the property of making 
the pigments gradually more translucent and deeper 
in tone. 

It is to these causes that the lowering of tone of 
pictures painted in oil is due. It is therefore essential 
that the artist should have a thorough understanding 
of how these changes are going to affect different pig- 
ments and different methods of painting, in order to 
avoid serious lowering of tone and serious changes in 
the whole colour scheme. 

The story told by Vasari of the discovery of how to 
paint in oil by the Brothers Van Eyck has misled artists 
for generations. So supreme a technical result so quickly 
arrived at, suggested a secret and lost medium, and 
Vasari himself hints at such a mystery. But when we 
realize that Van Eyck was the final expression of 
craftsmanship to produce a certain aesthetic result, with 



some three hundred years of experience and tradition 
behind him, we look rather to a study of his methods 
than of his medium for an explanation of his results. 

The main facts are, that, with the exception of an 
occasional obscure reference or isolated recipe, the 
account of how to grind pigments in oil and to paint 
in oil is essentially the same in the writings of Theophilus 
and Eraclius, in the twelfth century, in Cennino Cennini 
and other fifteenth-century manuscripts, and in Vasari 
in the sixteenth century. Chemists may dispute as to 
refinements, and the results to be obtained by different 
methods of preparing the oil described in these old 
recipes ; the extent to which resins were dissolved in 
the oil ; and what diluents were known. These details 
we may leave aside for our present purpose. 

The accumulation of evidence is in favour of the con- 
clusion that these painters were painters in oil, but 
probably on a solid under-painting in egg ; the extent 
to which this solid under-painting was carried being 
a matter for discussion. A study of these and later 
pictures, more especially those that are half-finished, 
reveals a supreme knowledge of the behaviour and 
properties of the dried oil film. No sharp line can be 
drawn between the perfection of preservation of the 
pictures of Van Eyck and his followers, and of some, 
at any rate, of the pictures of the later painters. More 
especially is this true of the Dutch school, at a time when 
the oil medium was undoubtedly firmly established. 

There can be no doubt that oil pictures painted on 
wood are much more brilliant and better preserved 
than pictures on canvas. 

The first step necessary for the student to-day is to 
learn something of elementary optics before he can 
understand how to handle the most pliable and, at the 
same time, the most treacherous of mediums. 




In the following chapters I have tried to bring together 
as shortly as possible such documentary information 
of which I am aware on early methods of oil painting. 
Before beginning the account of such evidence as I have 
been able to collect, it is, I think, necessary to state how, 
in my opinion, such evidence should be used. 

In the first place it is probable that the written 
evidence is both inaccurate and incomplete. The 
examination of the innumerable recipes published in 
early manuscripts compels one to the conclusion that 
in most instances the wTiters of the manuscripts were 
merely compilers, and had little practical acquaintance 
with the methods in use in the studio. Even when the 
writers are themselves painters, it is notorious that the 
account of technical processes written by those engaged 
in using them is often unreliable, and that essential 
details are omitted owing to the writers' familiarity with 
them, resulting in irritating obscurities. We must 
therefore take such information for what it is worth, 
and not regard it as conclusive. 

In the next place it is essential that this information 
be examined in a purely scientific spirit, without the 
making of unjustified assumptions. Unfortunately, too 
many writers on this subject have begun with a pre- 
conceived theory, and have had to twist historical 
evidence to prove a particular thesis. Let me give an 



example of how I think such evidence should be used. 

Dioscorides describes the preparation of nut and poppy 
oil. Oleum Cicinum was also known. These oils were 
therefore known in his time, but neither he nor Pliny 
mention their property of hardening into an elastic 
film, or their use as a medium for painting or the making 
of varnishes. This being so, until either written evidence 
is obtained, or the examination of contemporary works 
of art has proved the use of drying oils for such purposes, 
we must assume that in the time of Pliny the technical 
use of such oils for paint mediums and varnish was not 
known. Any assumption beyond this is mere specula- 
tion, and leads nowhere. 

So far as I know up to the present, no such use of 
drying oils has been found in the examination of objects 
of this or an earlier time. It is true that Greek phj'sicians 
mention the astringent properties of linseed, but are 
using the phrase in a medical sense, and it is also evident 
that a linseed poultice was kno^^^l in his time, but there 
is no mention of the extraction of linseed oil. 

The first mention of a use of a drying oil is made by 
^tius in the fifth centur3\ He describes the preparation 
of linseed oil, and after describing the preparation of 
walnut oil, states that it is used by gilders and encaustic 
painters to preserve their work owing to its property of 
drying. The first description of the preparation of an 
oil varnish, by dissolving resins in a drying oil, is found 
in the Lucca Manuscript, supposed to be of the eighth 
century, a recipe which is similar to that given in the 
Mappae Clavicula. 

It is not till we come to the manuscript of Theophilus, 
supposed to be of the eleventh or twelfth century, and 
the manuscript of Eraclius, supposed to be of about 
the same date, that we find an account of the use of 
a drying oil as a paint medium. 



Certain vegetable oils on exposure to the air in a 
thin film have the property of absorbing oxygen from 
the air and being turned into an insoluble elastic film. 
It is impossible to draw a sharp line between a drying 
and a non-drying oil ; but it is sufficient for our present 
purpose to note that poppy, walnut, and linseed oils were 
known as drying oils in mediaeval times, and these are 
the oils used by artists to-day. 

The oil as extracted from the seed or nut by heat, 
pressure, or by boiling with water, is far from pure, and 
contains mucilage and other impurities. Such an oil 
dries very slowly. It can be rendered more siccative by 
the following processes — boiling, passing air through it, 
and exposure to sun and air. All such processes also 
purify the oil, many of its impurities separating during 
the treatment. If the oil is repeatedly shaken up with 
water and exposed to sun and air, the mucilage and other 
impurities are separated, the oil is bleached and at the 
same time becomes more siccative. Such methods of 
purification are, we shall find, very old, and are practised 
by some artists' colourmen to-day. 

The oil so prepared dries quite sufficiently fast to be 
used for modern oil painting. There are also modern 
chemical methods of purification which are used for 
commercial oils. 

If it is required to hasten still further the rate of 
drying, the oil is boiled, or mixed in the cold and then 
exposed to light with certain metallic compounds such 
as litharge and white lead and compounds of manganese 
and cobalt. 

These compounds dissolve to a slight extent in the 
oil, and render it more siccative. 

An oil varnish is a solution of a resin in a drving 

These general facts about drying oils will enable us 



to follow the account which I now proceed to give of the 
old recipes. 

The bleaching of oils by the sun was known in the 
time of Dioscorides, and the boiling of oils with litharge 
was known in the time of Galen (a.d. 103-193). 

I shall quote in the first instance from the manuscript 
of Theophilus : — 

" Take linseed and dry it in a pan, without water, on 
the fire. Put it in a mortar and pound it to a fine 
powder ; then replacing it in the pan and pouring a 
little water on it, make it quite hot. Afterwards wrap 
it in a piece of new linen ; place it in a press used for 
extj^ acting the oil of olives, of walnuts, or of the poppy, 
and express this in the same manner. With this oil 
grind minium or vermilion, or any other colour you 
wish, on a stone slab, without water ; and with a brush 
paint over the doors or panels which you wish to redden, 
and dry them in the sun. Then give another coat and 
dry again. At last give a coat of the gluten called 
vernition, which is thus prepared." 

" Take any colours which you wish to apply, grinding 
them carefully in linseed oil, without water ; and 
prepare tints for faces and draperies, as you did before 
in water colours ; distinguishing, according to your 
fancy, animals, birds, or foliage with their proper 

" All kinds of colours may be ground in the same oil 
and applied on wood, but only on such objects as can 
be dried in the sun. For having applied one (coat of) 
colour you cannot add another until the first be dry, 
which in images (figures) and other paintings is too long 
and tiresome." 



*' All colours employed on wood, whether ground in 
oil or in gum (water), should be applied in three succes- 
sive coats. The painting being thus completed, place 
it in the sun, and carefully spread over it the gluten 
vernition. When this begins to flow with the (sun's) 
warmth, rub it gently with your hands. Do this 
thrice and then let it remain until it is tlioroughly 

" There is also a kind of painting on wood which is 
called translucid, or by some, golden ; it is produced 
as follows : Take a sheet of tinfoil — not varnished nor 
tinged with yellow, but in its natural state, and carefully 
polished — and line with it the surface which you wish 
to paint. Then having varnished the foil, grind colours 
very finely with linseed oil, and spread them extremely 
thin with the brush ; so let the work dry." 

These translations are from Eastlake's Materials for 
a History of Oil Painting. 

It is evident from these quotations that Theophilus 
knew how to prepare linseed oil, and how to use it as 
a painting medium. At the same time he gives no 
description of its purification, and if he did not know 
how to purify it, the oil would dry very slowly. His 
reference to the slowness of drying may be due to this, 
or it may merely be due to his being accustomed to 
paint with pigments mixed with egg and size, in which 
case oil would seem a very slow drying medium. 

The next quotation is from the manuscript of Eraclius, 
of which two copies are known, the oldest, formerly in 
the library of Trinity College, Cambridge, is of the latter 
half of the thirteenth century, and, therefore, the 
recipes are at any rate older than this date. These 
quotations are also from Eastlake's book, with one 
correction : — 



" If you wish to paint on a column, or on a stone slab, 
first dry it perfectly in the sun or by means of fire. Then 
take white, and grind it very finely with oil on a piece 
of marble. Spread the white with a brush two or three 
times over the column, which is (supposed to be) already 
quite smooth and even, without any cavities. After- 
wards prime with stiff white, applying it with your hand 
or with a brush, and let it remain a while. When it 
is tolerably dry, pass your hand with some pressure over 
the surface, drawing your hand towards you, and con- 
tinue to do this till the surface is as smooth as glass. 
You may then paint upon it with any colours mixed 
with oil. If 5^ou wish to imitate the veins of marble 
on a general tint (brown, black, or any other colour), 
you can give the appearance when the ground so pre- 
pared is dry. Afterwards varnish in the sun. 

" First plane the wood perfectly, rubbing the surface 
at last with shave-grass. If the wood is of such a nature 
that its roughness cannot be reduced, or if you have 
reasons for not wishing so to reduce it and at the same 
time are not desirous to cover it with leather or cloth, 
grind dry white lead on a slab, but do not grind it so 
finely as if you were to paint with it. Then melt some 
wax on the fire ; add finely pulverized tile and the 
white lead already ground ; mix together, stirring with 
a small stick ; and suffer the composition to cool. After- 
wards with a hot iron, melt it into the cavities till they 
are even, and then with a knife scrape away inequalities. 
And should you be in doubt whether it is advisable to 
mix the white lead with wax, know that the more you 
mix the harder it will be. The surface being smooth, 
take more white, finely ground with oil, and spread it 
thinly, with a brush adapted for the purpose, wherever 
you wish to paint ; then let it dry in the sun. When 
dry, add another coat of colour as before, rather stiffer, 



but not so stiff as to make it necessary to load the 
surface ; only let it be less oily than before, for great 
care is to be taken never to let the second coat be more 
fat (than the first). If it were so, and at the same time 
abundant, the surface would become wrinkled in drying. 
And now, not to omit anything that belongs to the sub- 
ject, I return to the first preparation of the surface of 
the wood. If, then, the panel on which you intend 
to paint is not even, cover it with leather made of horse- 
skin, or with parchment." 

Eraclius refers both to linseed and to nut oil. It 
will be noted that the panel is prepared with white lead 
ground with linseed and not with gesso. 

So far we have not found a recipe for purifying the 
oil for painting purposes, but we find in Eraclius the 
following recipe, translated thus from the Le Begue MS., 
by Mrs. Merrifield : — 

How Oil is Prepared for Tempering Colours 

" Put a moderate quantity of lime in the oil and 
heat it, continually scumming it. Add ceruse to it 
according to the quantity of oil, and put it in the sun 
for a month or more stirring it frequently, and note 
that the longer it remains in the sun the better it 
will be. Then strain and mix and keep the colours with 

This is not in the MS. at the British Museum and may 
have been interpolated. 

This method of preparation would result in an ex- 
cellent oil suitable for grinding colours, and for painting. 
It would dry somewhat faster than that used for grind- 
ing artists' colours to-day. The lime would dry the oil 
and neutralize any free acids and the white lead would 



render it siccative. We have therefore established that, 
at any rate as early as the thirteenth century, methods 
of preparing oil were known which would make an 
excellent paint medium, and that the grinding of pig- 
ments in this oil and its use were perfectly well known. 
There is no need to quote reproductions and variations 
on these recipes by later writers, and we shall now pro- 
ceed to the information given by Cennino Cennini on 
this same subject. 

Cennino Cennini is best known as a writer on tempera 
painting, but he gives us some useful information on 
how to paint in oil. Taking Lady Herringham's trans- 
lation on page 78, we find the following information : — 

How TO Paint in Oil on Walls, Panels, Iron, 


" Before we proceed further, I will teach you to paint 
in oil on walls, or on panels, which is much practised 
by the Germans, and in the same way on iron or stone. 
But we will first speak of walls. 

*' How to begin painting in oil on walls. 

" Cover your wall with plaster, exactly as you would 
do when painting in fresco, except that where you then 
covered but a small space at a time, you are now to 
spread over your whole work. Make your design with 
charcoal, and fix it with ink or verdaccio, tempered. 
Then take a little glue, much diluted with water — a 
whole egg, well beaten in a porringer, with the milky 
juice of the fig tree, is a still better tempera, you must 
add to the said egg a glassful of clean water. Then, 
either with a sponge or with a very soft brush without 
a point, go once over the whole ground on which you are 
going to paint, and leave it to dry for one day at least. 



" How to make oil jit for tempering colours, and also for 
mordants, by boiling over the fire. 

" It will be very useful to you to know how to prepare 
this oil, for many things that are done ; therefore, take 
one, two, or three, or four pounds of linseed oil, and put 
it into a new pipkin ; if it is glazed, so much the better. 
Make a small furnace, and make a round hole into which 
the pipkin fits exactly, so that the flame may not reach 
it, because the fire easily catches it, and there would be 
danger to the oil, and also of burning the house. When 
you have made your furnace, put a moderate fire in it, 
and the more slowly your oil burns, the better and more 
perfect will it be. Let it boil until it is reduced to half 
the quantity. But to prepare mordants, when it is 
reduced to half the quantity, add to each pound of oil 
one ounce of liquid varnish {vernice liquida), and let it 
be very fine and clear ; an oil thus prepared is good for 

" How to prepare good and perfect oil by cooking it in 
the sun. 
" When you have prepared this oil (which is also 
cooked in another way, better for painting, but not for 
mordants, for which it must be done on the fire, that is, 
cooked), take your linseed oil and in summer time put 
it in a basin of bronze or copper. And in August place it 
in the sun ; and if you keep it there ujitil it is half 
wasted, it will be exactly right for mixing with colours. 
And you must know that in Florence I have found 
the finest and best there can be. 

" How to grind colours in oil and to use them on walls. 

" Let us return to grinding the colours. Begin and 
grind colour by colour, as you did when working in 
fresco, except that where you then ground them with 



water, you must now grind them with oil. And when 
you have ground them, that is to say, all the colours 
(for every colour can be mixed with oil except bianco 
sangiovanni), provide small vessels, either of lead or of 
tin, into which put these colours. And if you cannot 
find such, get glazed vessels, and put the ground colours 
into them. Put them in a box that they may keep clean. 
When you would paint a drapery with three gradations 
of colour, as I have previously taught you, divide the 
space, and let each colour be laid in its proper place 
with a minever brush, uniting one colour well into 
another, the colours being very stiff. Then stop for a 
few days and return again to your work, see how the 
paint covers, and repaint where necessary. And in this 
way paint flesh or anything you please ; and in this 
way mountains, trees, and every other work. Provide 
a vessel of tin or lead (something like a lamp), about the 
height of your finger, half fill it with oil, and keep your 
brushes in it that they may not dry." 

In this account by Cennino Cennini there is only 
one obscurity, and that is his description of how to 
prepare a suitable oil for painting, by exposure to the 

In the recipe for preparing a mordant for laying on 
gold leaf, he says, " Let the w^hole be boiled till it be 
reduced to one-half" {'' E fallo bollire per mezzo e sta 
bene "). 

His recipe for preparing the oil for painting is usually 
translated, " and if you leave it so exposed till it be 
reduced to one-half." This makes nonsense of the whole 
paragraph, as no length of exposure to the sun would 
result in reducing the volume of oil in this way, and 
therefore we have to ask whether this is a correct 
translation, or whether there is some other explanation. 



The Italian, as quoted by Eastlake, is as follows, " il 
quale, se vel tieni tanto che torni per mezzo, e perfettissimo 
da colorirey 

The -word "mezzo," to-day, when used as an adjec- 
tive, may also mean faded, and there is a southern 
Italian phrase in which " mezzo " is used with the word 
for colour, which means something that is pale or 

I suggest, therefore, that the correct translation of 
this recipe is, " and if you keep it there until it is bleached, 
it will be perfect for colouring." 

Oil prepared in this way, although not drying so 
quickly as the oil prepared according to the recipe by 
Eraclius, would be perfectly satisfactory for the carrying 
out of oil painting. 

The next recipe of interest is from a manuscript in 
the Strassburg Library, of the fifteenth century, which 
is as follows (this translation is taken from Eastlake) : — 

How TO Temper all Oil Colours 

" Now I will also here teach how all colours are to be 
tempered with oil, better and (more) masterly than 
other painters ; and in the first place how the oil is 
to be prepared for the purpose, so that it may be limpid 
and clear, and that it may dry quickly. 

" How to prepare oil for the colours : Take the oil 
of linseed or of hempseed, or old nut oil, as much as you 
please, and put therein bones that have been long kept, 
calcined to whiteness, and an equal quantity of pumice 
stone ; let them boil in the oil, removing the scum. 
Then take the oil from the fire, and let it well cool ; 
and if it is in quantity about a quart, add to it an ounce 
of white copperas ; this will diffuse itself in the oil, 
which will become quite limpid and clear. Afterwards 



strain the oil through a clean linen cloth into a clean 
basin, and place it in the sun for four days. Thus it 
will acquire a thick consistence, and also become as 
transparent as a fine crystal. And this oil dries very 
fast, and makes all colours beautifully clear, and glossy 
besides. All painters are not acquainted with it ; from 
its excellence it is called oleum preciosum, since half an 
ounce is well worth a shilling ; and with (this) oil all 
colours are to be ground and tempered. All colours should 
be ground stiffly, and then tempered to a half Hquid 
state, which should be neither too thick nor too thin. 

" These are the colours which should be tempered 
with oil : vermihon, minium, lake, brasil, red, blue 
bice, azure, indigo, and also black, yellow, orpiment, 
red orpiment, ochre, face brown-red, verdigris, green 
bice, and white lead. These are the oil colours and no 
more. Here observe that these colours are to be well 
ground in the oil and at (last) with every colour mix 
three (that is, a few) drops of varnish, and then place 
every colour by itself in a clean cup, and paint what you 
please. With all the above-mentioned colours a small 
quantity of calcined bone may be mixed, or a little 
white copperas about the size of a bean, in order to 
make the colour dry readily and well." 

This recipe introduces us to two new processes. 
One, to use white copperas as a drying agent instead 
of white lead. I have not found in my experiments 
that pure dehydrated sulphate of zinc has any value as 
a drying agent, but it has been suggested that the 
impure sulphate of zinc available in mediaeval times 
would probably contain manganese, and therefore act 
as a drying agent. 

The other new method is the addition of a few drops 
of varnish to the colour after it is ground. 

C 33 


I have not troubled here to quote the many account- 
rolls which exist in this country, showing that in the 
thirteenth and fourteenth centuries painting in oil was 
common, as these accounts merely confirm what we 
have already learned, and they are discussed in the 
chapter on Wall Painting. 

We have now brought our account up to and well 
into the fifteenth century, and it will be noted that there 
is no indication of any marked change of practice or any 
historical evidence for the account given by Vasari of 
the discovery of a new technique by Van Eyck. I 
therefore omit Vasari's description of Van Eyck's 
discovery of a new oil technique which has probably 
no historical basis, and which is too vague to give us 
any useful information. 

Vasari, however, gives a very definite account of 
the technique of his own time which is as follows, and 
which is quoted from Vasari on Technique, by Louisa 
Maclehose : — 

" I must now explain how to set about the work. 
When the artist wishes to begin, that is, after he has 
laid the gesso on the panel or framed canvas and 
smoothed it, he spreads over this with a sponge four or 
five coats of the smoothest size, and proceeds to grind 
the colours with walnut or linseed oil, though walnut oil 
is better, because it yellows less with time. When they 
are ground with these oils, which is their tempera 
(medium), nothing else is needed so far as the colours 
are concerned but to lay them on with a brush. . . . 
' Vanyio poi macinando i colori con olio di noci o di 
seme di lino {benche il noce e meglio perche ingialla meno) 
et cosi macinati con quesii olii, che e la tempera loro, 
non bisogna altro quanto a esii, che distenderli col 
penello.'' " 



I give the Italian here as Professor Berger suggests 
that this translation means that Vasari used an emulsion 
of egg and oil. 

Recipes for purifying oil by exposing to air and sun 
over water come somewhat later. Eastlake quotes one 
from the compendium of Padre Gesuato (1557), and 
De Mayerne (1620) gives more than one recipe. The 
only point to be noted here is that the older recipes 
I have quoted, in which lime is added, would produce 
a neutral oil, while an oil purified by washing with water 
and exposing to air and sunlight is very acid. 

It is evident then that from the twelfth century 
onwards the methods of preparing and purifying oils 
were much the same as those practised to-day, and 
there is no indication of any secret processes which 
would give a different result to those with which we are 
familiar. It will also be noted that in the earlier recipe 
no distinction is made between one drying oil and 
another, but that Vasari tells us distinctly that linseed 
oil darkens and that nut oil does not darken so badly. 
I do not think sufficient attention has been paid to this 
remark by Vasari. 


The recipes for varnishes are so innumerable that it 
would be impossible to quote them, but we can divide 
them into two groups : the recipes before 1500, and 
the recipes of the sixteenth century and later. 

To deal first with the recipes before 1500. These are 
all recipes for dissolving resins in oil, spirit of turpentine 
and alcohol being commercially unknown. While it 
is difficult to identify the resins referred to, the pine 
balsams, Venice turpentine, and olio d'abezzo, rosin, 
mastic, sandarac, and possibly amber, seem to have 
been used, either separately or mixed together, by 



dissolving in boiling oil. The recipes for amber are of 
three kinds : either the amber is first dissolved in an 
olio resin like Venice turpentine, in which case it would 
not have the special properties of a pure amber varnish ; 
or the amber is fused and then dissolved in oil, which 
gives a dark and very slow- drying varnish ; or the 
amber is added to the boiling oil without fusing, in which 
case it does not dissolve. 

It seems very doubtful if varnish made from amber 
was used in practice. The varnishes made from the 
other resins, judging by the recipes, contain a much 
higher proportion of resin to oil than would be used 
to-day. They would not be regarded by the modern 
varnish-maker as nearly as durable as varnishes made 
from copal. They were not thinned with turps, like 
the modern oil varnish, were very sticky and thick, and 
were rubbed on warm with the hand. 

In the sixteenth century recipes are given for the 
first time for spirit varnishes, in which the resins are 
dissolved in spirit of turpentine or alcohol — correspond- 
ing therefore to modern spirit varnishes, such as mastic 

Evidence for the Use of Emulsions 
It remains next to consider what evidence there is 
for the use of emulsions of egg and oil or egg and varnish. 
Vasari, in his Life of Alessio Baldovinetii, says that he 
experimented with emulsions, yolk of egg and " vernice 
liquida," and found it unsatisfactory as the colours 
cracked and peeled off when applied too thickl}- ; and 
he refers again to such experiments in his Life of Anto- 
nello da Messina. 

Again, in the Venetian Manuscript in the British 
Museum, written probably in the first half of the 



fourteenth century, there is a recipe for painting on 
glass by using a mixture of yolk of egg and varnish. 

The following is an interesting recipe for a wax 
emulsion given by Le Begue : — 

" If you wish to make a water proper for distempering 
all colours, take one pound of lime and twelve pounds 
of ashes ; then take boiling water and put the whole 
together, making them boil well ; after which let the 
mixture settle and strain it through a cloth ; then take 
four pounds of that water, heat it well, take about two 
ounces of white wax and put this to boil with the water. 
Then take about one and a half ounces of fish glue, put 
it in water and leave until it is well softened and — as it 
were — melted, when you must manipulate it until it 
becomes like paste, and throw it into the water with 
wax and make all boil together. Then add to it about 
an ounce and a half of mastic, and boil it with the other 
ingredients. Take some of this water on a knife blade, 
or piece of iron, to ascertain whether it is done. If it is 
like glue, it is all right. Strain this water while hot or 
tepid through a linen cloth, let it settle, and cover it 
well. With this water you may distemper all kinds of 

A similar recipe is given in the Mount Athos MS. 

In the Bolognese MS. of the fifteenth century, trans- 
lated by Mrs. Merrifield, a recipe is given for grinding 
w^iite lead, lake, linseed oil, and white of egg together ; 
and later in the same MS. a recipe for a cement made of 
white lead, varnish, and white of egg. 





Another matter which requires investigation is how 
far by exposure to bright dayhght or sunhght the 
tendency of an oil film to darken can be permanently 

That such a darkened film can in most cases be 
bleached by such exposure we know, and there is more 
than one reference to the exposure of oil pictures to 
sunlight in the history of painting. Of these the follow- 
ing from a letter written by Rubens is, I think, the most 
interesting : — 

" If I knew that my portrait was still in Antwerp, 
I would have it detained there, to have the box opened, 
to see if it has not been injured, or become darkened, as 
happens often to fresh colours, if they are, as is here the 
case, so long in a locked box, and not in contact with the 
air. It may be then that my portrait does not now look 
as it did originally. Should it really reach you in such 
a bad condition, the best remedy for that would be to 
put it often in the sun ; by this means the excess of oil, 
which causes such changes, is destroyed ; and if, from 
time to time, it should again get dark, setting it in the 
sun's rays must be renewed. This is the only remedy 
against this heart disease." 



It is evident from this quotation that the pictures 
as they left the studio of Rubens were not in such a 
condition as to resist the darkening of the oil if rolled 
up. If, therefore, he had given them a sun bath, it had 
not been sufficient permanently to bleach the oil, if that 
is possible. It is evident also that he does not regard 
such a bleaching process as producing a permanent 
result, as he says that if the picture again darkens, it 
must again be placed in the sun. 

This passage has been quoted as proving that it was 
the practice of painters in the time of Rubens to place 
their pictures in the sun as part of their painting 
method, and in this way the tendency of the oil to 
darken was permanently cured. It is, I think, evident 
from a careful analysis of this passage that the 
evidence to be derived from it tends rather the other 

It is nevertheless a matter for experiment to find how 
far sun-bleaching of the dried film affects a permanent 
cure of the tendency to darken. Professor Eibner has 
shown that exposure to sunlight while drying is most 
injurious to an oil film. 

The two most valuable treatises on the history of 
painting are Eastlake's Materials for a History of Oil 
Painting, a model of scholarship and research ; and 
Professor Berger's Beitrage zur Entwickelungs-Geschichte 
der Maliechnik, in which the De May erne MS., quoted 
by Eastlake, is translated in full. Translations of 
individual manuscripts by Mrs. Merrifield and Lady 
Herringham, and Vasari on Technique, by Louisa 
Maclehose, can also be consulted. 

Having quoted the more important recipes given in 
the old MSS., it is necessary next to consider critically 
the conclusions come to by Professor Berger and by 



At the time when Eastlake wrote, the story of the 
discovery of oil as told by Vasari still had general 
acceptance, and no doubt influenced Eastlake through- 

Modern critical examination of contemporary records 
about the Brothers Van Eyck have not resulted in pro- 
ducing any evidence as to the truth of this account, 
and it is certainly remarkable that no mention was made 
of it on the tombstone of either painter, both of which 
had inscribed upon them lengthy epitaphs. 

Eastlake suggests that the only oil painting known 
before Van Eyck was in a thickened oil obtained by 
exposure of the oil to sun and air. If the crude oil is 
purified by the process so often repeated in old recipes 
of exposure to sun and air, over water, the first change 
is the precipitation of the mucilaginous and albumenoid 
impurities. This is followed by three changes. The oil 
bleaches and at the same time is partially oxidized and 
partially polymerized. The partial oxidation makes it 
dry more quickly ; and the polymerism will thicken it in 
time. The rate at which these two processes will take 
place is no doubt influenced by temperature and light 
acting simultaneously, and so different results will be 
obtained under different conditions. 

In practice, at any rate in this climate, there is no 
difficulty in preparing an oil, pale, limpid, and suffi- 
ciently quick drying for ordinary artists' purposes. 
I worked out the conditions some thirty years ago, and 
such an oil has been used commercially for grinding 
artists' colours ever since without the addition of any 
dryers. Eastlake is wrong, therefore, in assuming that 
an oil could not be prepared in this way suitable for 
painters. If the process is continued long enough the 
oil will thicken and will then be more suitable for use 
as a varnish. 


the Tf 

: .\utitinai 


srees ty - 

HOLY FA5IILY. By the Master of " The Death of the Vibgix." 
(Xational Gallerr, London.) 

This picture has been selected in preference to the Van Eycks in the Xational Gallery, because it 
is in even better preservation than they are, and is a perfect example of the triumph of the 

technical methods employed by the early painters in the tempera oil technique. 

The picture seems as perfect as when it left the artist's studio. For further information the 

reader is referred to Chapter XI. 


In the Hermeneia, instructions are given for preparing 
an oil by exposure to the sun till it is as thick as honey, 
and in a fifteenth-century recipe quoted from the Stras- 
burg Manuscript, it is stated that after three days* 
exposure to the sun the oil boiled with bone ashes and 
treated with sulphate of zinc becomes thick. The 
thickening must have been due to the boiling, not to 
the short exposure to the sun ; and if so, the boiling 
must have been prolonged. If this recipe is to be taken 
as conclusive, then in the fifteenth century they ground 
in a thickened oil, that is, polymerized oil like Stand 

The other evidence is supposed to be given by Cennino 
Cennini. This, I believe, is due to a mistranslation, as 
I have already stated. Cennino Cennini describes two 
methods of preparing linseed oil — one by prolonged 
boiling, for making mordants for cementing on gold 
leaf; and another, which he says is the better way for 
a painting oil, by exposure to the sun and air. As the 
second recipe makes nonsense as it is usually translated, 
I have suggested the alternative translation that the oil 
is exposed until it is bleached. 

In this connection there is a passage in Cennino 
Cennini, Chapter 144, which seems to have escaped 
attention : — 

" If you would imitate velvet, paint the drapery of 
any colour you please, tempering your colour with the 
yolk of an egg. Make the pile on the velvet with a 
minever brush tempered with oil ; shape the pile with 
thick paint." 

This instruction surely shows a manageable oil 
medium, and it is difficult to understand why, having 
a method by which a satisfactory painting oil could be 



produced, the painters previous to Van Eyck should 
have gone out of their way to make it too thick. 

This question of the use of thickened oils has recently 
received renewed attention because of the interesting 
suggestion of Professor Eibner that Van Eyck and his 
followers used Stand oil, i.e. linseed oil thickened and 
polymerized by prolonged boiling in absence of air, and 
also possibly by prolonged exposure to the sun's rays 
without oxidation. There are chemical reasons for 
stating that such an oil would produce a very permanent 
dried film which would not yellow so much as oil pre- 
pared in the usual way, and some of the old recipes 
would give a partially polymerized and partially oxidized 
oil. Professor Eibner's view differs from that expressed 
by Eastlake, in that he suggests the use of thickened 
oils for the perfected, not for the imperfect, method of 
oil painting. I have tried grinding pigments and 
painting in this thickened oil. It is sticky and at the 
same time flowing. Fine work like that of the fifteenth- 
century painters is impossible with such a medium. If 
used, the oil was either thinned with oil of turpentine or 
emulsified with a little egg. Professor Eibner is of the 
opinion that Van Eyck and his followers must have 
known and used oil of turpentine or a similar volatile 
medium, the use of which requires a knowledge of the 
art of distillation. 

The earliest reference to distillation is made by Pliny, 
who speaks of the distillation of a crude turpentine by 
heating pine balsam in a vessel covered with a woollen 
fleece and squeezing out of the wool the condensed 
vapours. Distillation, as we know it, is first described 
by Cleopatra, a physician of the third century in Egypt, 
and must have been a familiar process to the alchemists, 
and the preparation of turpentine is described by Marcus 
Graccus in the eighth century. That Van Eyck and his 



followers could have used oil of turpentine must there- 
fore be admitted. 

In the Advocates' Library, Edinburgh, there is an 
illuminated manuscript which is known to have been 
painted between 1465 and 1489, and in which wax has 
been used as a medium. Pure beeswax has been used, 
as I have proved by taking the melting point of the 
substance, so that it is not an emulsion of wax and glue, 
such as is described by Le Begue, and the delicate 
painting could only have been done by dissolving the 
wax in a volatile medium like turpentine. 

This then gives us a date when turpentine must have 
been known. 

It is not likely that this is the only manuscript in 
which pure wax has been used, and a search for others 
would enable us to arrive at a conclusion as to when 
turpentine was knowTi as a medium for artists. 

In the meantime the silence of manuscripts before the 
sixteenth century is against Professor Eibner's view. 
Sixteenth-century manuscripts contain numerous recipes 
for the making of spirit varnishes by dissolving resins 
in volatile mediums. None are mentioned of an earlier 

The other possibility is the emulsifjdng of thickened 
oil with egg. There are two ways in which an egg 
emulsion can be used. By the addition of the white, but 
still better with a little of the yolk of egg, to a sticky 
varnish or sticky balsam, or sticky thickened oil, it is 
possible to obtain a medium which is crisp and will 
not flow. By the addition of more egg and water it is 
possible to obtain a medium which will mix with water 
and paint out like a size medium and dry dead and flat. 
All this has been dealt with in detail in the chapter on 

The theory that egg emulsions miscible with water 



were used by Van Eyck and his followers, has been 
developed at great length and with great ingenuity by 
Professor Berger, and it is necessary, therefore, to look 
at the actual evidence. 

There are, I believe, only three references in ancient 
writing. One is the reference in Vasari, where he states 
they were tried and abandoned. Another is the recipe 
in the Bolognese manuscript of the fifteenth century for 
grinding white lead, linseed oil, and white of egg together, 
and the third, in the Venetian Manuscript in the British 
Museum, of the first half of the fourteenth century, is 
for grinding yolk of egg and varnish together for painting 
on glass. 

There is no evidence that in either of these two recipes 
water was also to be added. On such slender evidence 
does this elaborated theory as to fifteenth-century 
technique rest. 

Against these occasional recipes we have to place the 
evidence as to the method of painting in oil, which is 
so often repeated in different manuscripts. In Theo- 
philus, in Cennino Cennini, and in Vasari we find the 
same simple instruction repeated, to grind the pigments 
in linseed or walnut oil, and this, as Vasari tells us, is 
their only tempera. 

Eastlake, having done his best to establish an improve- 
ment made by the Brothers Van Eyck on the methods 
of oil painting, comes to the conclusion that their secret 
lay in the use of an " oleo resinous vehicle." 

As has been already seen, the mediaeval varnishes 
were thick, sticky, and slow drying, and quite unsuitable 
for painting. There is evidence that the addition of 
a few drops of varnish was made to the oil medium, 
both in the recipe already quoted from the Strassburg 
Manuscript, and also in the instructions given by later 
writers on painting. Such an addition would give a 



smoother, harder, and more polished surface than oil 
alone, and it may very well have been the praetice to 
add a little varnish in this way. Such an addition is 
often made by modern painters, but it can hardly be 
claimed that so small a modification would profoundly 
affect the durability and brilliancy of painting in oil. 

The whole question can be approached in another 
way, and that is the consideration of the pigments which 
were available, and how far these pigments were perma- 
nent or fugitive, and whether, if fugitive, anything could 
be done to protect them. 

Many years ago I published a method of discovering 
how far oil and varnish films were permeable to moisture 
by grinding them with ignited sulphate of copper. 
Sulphate of copper, or blue vitriol, crystallizes in trans- 
parent blue-green crystals together with five molecules 
of water, which can be driven off by heat, leaving an 
opaque white powder. If this opaque white powder 
is ground with linseed oil, a white enamel is obtained, 
and if after drying in dry air it is exposed to water 
vapour, owing to the easy permeation of the linseed oil 
film by water, the sulphate of copper combining again 
with water is converted into a transparent bluish green. 
Experiments on these lines showed that linseed oil, 
w^alnut oil, and the usual oil varnishes were easily per- 
meated, but the balsams and pure resins dissolved in 
a volatile medium resisted for a long time the attack 
of water vapour, and therefore it is possible to protect 
fugitive pigments from change by locking them up in a 
balsam or a varnish made by dissolving resin in a volatile 
medium, and it was also found that a very small addition 
of oil could be made to such a balsam without spoiling 
its protective value. I therefore suggested that 
possibly in the case of fugitive pigments such balsams 
as Venice turpentine or Oleo d'Abezzo were used. Since 



then quantitative methods for testing both the absorp- 
tion of water by oil films and their permeability have 
been devised, but the main facts remain as I have already 
stated them. 

When engaged in the examination of the pigments 
used on illuminated manuscripts, I found a green which 
apparently is a resinate of copper made by dissolving 
verdigris in a pine balsam and which is permanent. 
If I am right in supposing that this green was used by 
Van Eyck and his followers, it may have been used by 
taking the sticky solution of the green in the pine balsam 
and emulsifying with a little egg and oil so as to make 
a workable medium. Such a mixture is opaque when 
first painted out, but turns transparent on drying. Or 
the green balsam may have been dried and ground and 
used as a pigment, so that even if we take the use of 
this green as proved, it does not prove the use of a 
balsam varnish in the actual painting. 

If we now proceed to consider the pigments available 
in the fifteenth century, unless we find that fugitive 
pigments were the only obtainable pigments, there is 
no reason for us to suppose that any locking-up process 
was essential. Very large numbers of pigments are 
described in the various manuscripts, many of which 
are notoriously fugitive, but the result of my examina- 
tion of illuminated manuscripts was to show that in 
most cases at any rate a very limited palette was used — 
white lead, vermilion, red lead, orpiment, malachite, 
copper resinate, verdigris, possibly gamboge, real ultra- 
marine, azurite, and a lake. 

For the artist's palette painting in oil we must add 
to this list the yellow oxide lead or massicot, Naples 
yellow, the oxide of lead and antimony, and probably 
the oxides of lead and tin, and, of course, the earth 
colours and the blacks. The oxide of lead and the 



oxides of lead and antimony have been removed from 
the artist's palette on account of the fear of their turning 
black owing to the action of sulphuretted hydrogen ; 
but the durabihty of white lead makes it doubtful if 
this removal is justified. If we remove verdigris from 
the list, replacing it by the resinate of copper, the 
probability is that we can regard this as a permanent 
palette, with the exception of gamboge and the possible 
exception of orpiment. Sir Arthur Church condemns 
orpiment as fugitive on account of some experiments 
made by Sir Joshua Reynolds, which incidentally confirm 
the value of a pine balsam as a preserving agent, but the 
probability is that the orpiment he used was an artificial 
preparation. In the old manuscripts we find that orpi- 
ment is stated to be a bad drier, and that it must not 
be mixed with pigments containing lead or copper ; but 
in no case is it spoken of as fugitive, and it has proved 
to be quite permanent when used on illuminated 
manuscripts. It has therefore, I think, been too hastily 
condemned. I have, however, found one example, in 
a picture by Holbein, of the use of powdered gold and 
yellow ochre in its place, which looks as if he had 
regarded it with suspicion. 

With reference to the copper carbonates, malachite, 
azurite, and the artificial varieties blue and green bice, 
there is sufficient evidence on old pictures as to their 
permanency as pigments, the only objection to them being 
their tendency to settle in the oil, and owing to their 
transparency to suffer excessively from the yellowing 
of the oil, though in some cases they have turned black. 

There still remains for us to consider what lakes were 
available. The lakes were prepared from Japan wood, 
kermes and later on cochineal, Indian lac, ivy gum, and 
madder. All of these, with the exception of madder, 
are fugitive and recipes fos madder are remarkable by ' 



their absence ; but we know that madder was used as 
a dye, and there are many recipes for extracting the 
colour from dyed wool to make lakes ; while the per- 
manency of many of the lakes used not only in oil but 
in tempera is evidence in favour of madder lakes being 

Finally, the good preservation on the whole of the 
pigments in tempera pictures, which have no such pro- 
tection as is afforded even by oil, and which, even if 
the pictures were originally varnished — which I venture 
to doubt — have long lost their protective varnish, tends 
to prove that within a limited range the fifteenth- 
century painter had a permanent palette, and there- 
fore there was no need for him to resort to the use of 
pine balsams or of highly resinous varnishes to protect 
his pigments. I frankly confess that I am here express- 
ing a view which is different from what I have ex- 
pressed before, but on revising the whole of the evidence 
available, it seems to me that the verdict is in favour 
of the use of the oil medium from the Middle Ages on- 
wards, with the pro^dso that a little varnish may have 
been added, and also that probably before the intro- 
duction of turpentine as an artist's medium, when an 
artist wished to use a sticky oil or varnish, he mixed it 
for this purpose with a little egg. 

If this \iew is correct, we are driven to the conclusion 
that the fine condition of the early Flemish pictures and 
their high colour key after the lapse of so many cen- 
turies, is due in the first place to very careful and 
methodical painting on a sound wood panel, and in the 
second place to the thorough understanding and success- 
ful use of the combined tempera oil technique. 

The next question of interest which arises is how and 
when the combined oil tempera technique began. The 
oldest Northern tradition, as given by Tbeophilus and 



Eraclius, is to prime the panel with white lead in oil and 
paint with oil. If we go still further back to the Lucca 
Manuscript, we find painting in wax. Vasari, in order 
to heighten the dramatic effect of his story about Van 
Eyck, tells us that before his discovery, egg was the 
ancient and universal medium, and this statement has 
been copied by every subsequent writer on art, but I 
cannot find that it is based on actual evidence. East- 
lake tells us that in certain pictures he examined of the 
fourteenth century, the draperies were painted in oil 
and the faces in tempera ; and gives as his evidence 
that the flesh painting is thinner than the painting of 
drapery. We now know that even sixteenth-century 
oil painters laid thin paint on the flesh, in order to obtain 
full advantage of the white gesso below, so that these 
pictures to which he refers may well have been oil 
pictures throughout. 

It is very difficult to tell by inspection in what medium 
pictures previous to 1400 were painted, and a chemical 
investigation of the medium used should yield results 
of extraordinary interest. 

Were egg and wax the two classical mediums, and 
does the painting in egg continue throughout the 
centuries, side by side, Avdth the painting in wax which 
somewhere between 800 and 1200 was replaced by oil ? 
Is the Flemish miracle of the fifteenth century the com- 
bination of the two parallel methods for the first time 
in the hands of a master — not as Professor Berger would 
have us believe, by mixing egg and oil on his palette, 
but by utilizing the full optical advantages of both 
methods in a combined technique ? It is in a chemical 
examination of the pictures painted in the North before 
the Van Eycks that an answer to these questions will 
be found. 

In the follomng chapters we shall discuss artists' 

D 49 


materials, paper, wood, canvas, pigments, and oils, and 
the optical principles involved in oil painting and the 
optical properties of oils and pigments. 

We shall then be in a position to discuss early ex- 
amples of oil technique, and throw further light on the 
methods used and the scientific principles underlying 




The earliest form of paper known in Europe is that 
prepared from the pulp of the papyrus reed in Egypt. 
This was used during the time of the Roman Empire, 
and seems ultimately to have gone out of use and been 
replaced by vellum, owing to the exhaustion of the 
supply of the papyrus reed, and consequently we are 
accustomed to find vellum used for early manuscripts. 

The manufacture of linen paper has not been traced 
further back than the second half of the twelfth century. 
Specimens of thirteenth-century paper exist, and are 
in excellent condition. They have been sized with 
starch, the size used by the Egyptians in preparing 
papyrus. To-day paper is usually sized with size. 

Paper for water-colour painting should be a rag paper, 
preferably prepared from linen fibre, and should con- 
tain nothing in addition but a little size, all such chemi- 
cals as bleaching powder, sulphite of soda and alum 
being absent. In a recent analysis of a high-class 
water-colour paper, I found that the paper consisted of 
linen fibres with a little cotton fibre, and Avas quite free 
from any oxidizing or reducing agents, whilst the amount 
of ash was from '56% to -71%, and consisted partly of 
soluble sulphates and of calcium sulphate. 

In the preparation of paper, the rags, having been 
boiled and thoroughly pulped, are usually bleached by 
the introduction of a little bleaching solution, the excess 
of the bleach destroyed by means of sulphite of soda, 



the pulp thoroughly washed, and some size incorporated. 
This size sometimes contains a little resinate of soda and 
sometimes alum is also added. In the case of water- 
colour paper, resinates and alum should be absent, as 
they may act on some of the more sensitive pigments. 
It is also very important that care should be taken that 
no fragments of iron or bronze have been introduced 
in the process of manufacture. This is most apt to 
happen from disintegration of the teeth of the beaters, 
and I understand that in making some of the finest 
qualities of paper in Germany, the teeth of the beaters 
are made of agate. These tiny particles of iron or bronze 
will oxidize in time, producing spots upon the paper, 
and will decompose certain pigments — cadmium yellow, 
for instance, being decomposed, forming black spots 
visible to the eye from tiny particles of iron or bronze, 
which would only have been visible in the original 
paper under the microscope. 

It is to be regretted that it is not possible to obtain 
a modern water-colour paper in which such powerful 
bleaching agents as bleaching powder have not been 
used. The old paper on which the dra-vAings of Raphael 
and other artists were made, and which has stood so 
well the test of time, must have been bleached simply 
by sun, air, and water. We have to-day a bleaching 
agent, which was doubtless produced under those con- 
ditions in small quantities, namely, peroxide of hydro- 
gen. This, strictly speaking, should be the only bleaching 
agent used in the preparation of water-colour paper. 
The question of the amount of size in the paper, and 
consequently the extent to which it is non-absorbent, 
is one of great interest to the water-colour painter. It 
is fully discussed by Mr. Rich in his book in this series 
on Water-colour Painting. 



Wood Panels 

Well-seasoned panels of wood form an excellent sup- 
port for pictures. At the same time wood has certain 
obvious defects ; it tends to warp and crack, and is apt 
to be attacked by the wood beetle. It is necessary, 
therefore, in the first place to select a suitable variety 
which is least hkely to warp or crack. It should be free 
from resin and gums, and wood gi'own on poor soil in 
temperate climates and felled in winter is the best. 
In Italy the white poplar was largely used, while the 
Flemings used oak. To-day mahogany is used, and is 
a most satisfactory wood for the purpose, being much 
less likely to crack than oak. 

In order to understand the seasoning process through 
which wood has to go, it is necessary to know something 
of the microscopic structure of wood. Wood consists 
of cells which in the greenwood are more or less full of 
water, and the lining of these cells also consists of 
material by which water is absorbed, and is more or 
less loosely combined, and which is able to take in or 
give up water according to the amount of moisture in 
the air. 

Starting from the central pith, a fresh layer of wood 
is formed each year surrounding that of the previous 
year and lying underneath the bark, and so year by 
year a fresh ring is added to those already formed. As 
this process continues the inner or older rings cease 
to constitute part of the living portion of the wood body 
of the tree, becoming a central column of support, and 
acting as a storehouse for water. Surrounding this 
central heart-wood, which is sometimes darker in 
colour, is a zone of wood consisting of a varying number 
of year rings, usually lighter in colour than the heart- 
wood, and is called the sap-wood. This sap-wood is 



still alive, and takes part in conducting water from 
the roots to the leaves. Between the outermost year 
ring and the bark is a ring of actively growing tissue 
called the cambium. 

It is evident from this account that when the trunk is 
cut tangentially into planks, the condition of the cells 
is different on the two sides of the planks, and conse- 
quently, from the difference in the condition, there is 
a tendency to warp. The best way to cut timber to 
avoid warping is radially. In most varieties of timber 
the warping takes place almost entirely at right angles 
to the direction of growth. 

We may have warping, therefore, due to the different 
properties of the cells on each side of a panel, and due 
to difference in the amount of moisture in the air on 
the two sides of the panel. Prolonged and careful 
seasoning diminishes the danger of warping from both 
these causes, and the thoroughness of the seasoning can 
be tested by a method shortly to be described. What 
is known as ply-wood is prepared by cutting spirally 
round and round the tree a thin shaving and then 
glueing the shavings together with casein, with the 
grain of each piece at right angles to the last piece. Ply- 
wood differs very much in quahty and tendency to warp 
according to the care with which the wood has been 
seasoned and the kind of wood selected. 

When the wood is allowed to dry, the cell water first 
disappears, and then the cell walls begin to lose water 
more and more slowly. They never lose the loosely 
attached water completely until the wood is heated to 
a temxperature at which it begins to decompose. The 
seasoning, therefore, of timber is the reduction in the 
amount of water it contains to an average, and as long 
as this average is sustained afterwards the wood does 
not warp or crack. 



The process of seasoning is a slow one, and after 
seasoning in bulk and being cut up it requires to be 
seasoned again. There are now rapid processes in use 
for seasoning Anth the assistance of hot air and steam, 
but they have yet to be proved to be reliable for picture 
panels. Cennino Cennini recommends boiling in water 
and then drying afresh. Sir Arthur Church advises 
soaking in water heated to 50° C. and then steaming. 
When dry he washes both sides with a solution of cor- 
rosive sublimate and again dries and seasons in a warm 
air chamber, and finally protects the back of his panel 
with oil paint. 

As I have stated, wood is never dry. It retains 
a certain amount of moisture, and, in the presence of 
dry air, gives off moisture and shrinks and, in the 
presence of moist air, takes up moisture and expands. 
With well-seasoned wood kept under proper conditions, 
actual cracking from shrinkage of the whole of the wood 
should be impossible, but the wood is necessarily not 
exposed to the same conditions on both sides when 
hanging on the wall of a gallery, and, consequently, it 
■v\ill warp more or less, and the warping ^^i]\ varj'^ from 
hour to hour and day to day, as the conditions of 
moisture on the two sides of the panel change, a warping, 
which even if not perceptible, is loosening the hold 
between the priming and the wood. To get more 
information about this, I made the follo-w-ing experi- 

The panels tested were 12 in. each way, and I had a 
tin box made 11 1 in. by 11 1 in. and 4 in. deep with 
a broad brass flange attached around the top. I then 
took a sheet of rubber 14 in. each way, cut out the 
centre, and cemented the rubber strip on to the edge 
of the panel, just as one would mount a water-colour 
picture on to a cardboard mount. The panel with this 



rubber strip was then laid on the top of the box, another 
brass flange laid on the rubber and the two flanges 
pressed together with brass screw clips. The box was 
tlius closed by a lid, the panel, the rubber margin, and 
tlie brass flange resembling a picture, mounted on a 
mount, and surrounded by a frame and the wood was 
free to expand, contract, and warp in any direction. 
The diagram makes the whole arrangement perfectly 




Fig 1. 

Water was put inside the tin box so that the lower 
surface of the panel was exposed to air saturated with 
moisture, while the upper surface was exposed to the 
air of the room. After exposure for about twenty-four 
hours to moisture the panel was taken out, placed on a 
sheet of plate glass, and the amount of warp that had 
taken place measured with a steel rule. Readings with 
a wet and dry bulb thermometer placed above the box 
showed no very great change in the amount of moisture 
in the atmosphere during the experiments. 

The first experiment was made with a white wood 
panel supplied by a reputable artists' colourman, and, 



it may be taken as consisting of what is regarded as 
properly seasoned wood. The amount of warp amounted 
to 12 millimetres or approximately | in., the thickness 
of this panel being 5 millimetres. 

The next experiment was -snth a mahogany panel, 
3 millimetres thick, cut from a piece of mahogany which 
had been cut and seasoned for some three years. The 
amount of warp in this ease amounting to 9 miUi- 
metres. As this panel was thinner and therefore should 
have warped more than the white wood panel, this 
showed superiority of the mahogany over the white 

The next experiment was made with a mahogany 
panel also 3 miUimetrcs thick, and cut from an old piece 
of second-hand mahogany, at least twenty years old. 
The warp in this case only amounted to 5-5 millimetres, 
showing the superiority of this old mahogany and 
therefore the length of time that is required for a piece 
of wood to become thoroughly well seasoned. It is 
usually stated in the text -books that the seasoning of 
wood is merely a matter of the drying out of sap, but 
I do not think this is all the story. Apparently by long 
exposure to air the wood cells become less sensitive to 
changes in the moisture of the atmosphere, losing prob- 
ably their hygroscopic properties. 

My third experiment was made with a piece of three- 
ply wood 5 milHmetres thick, suppUed by the Venesta 
Company and made from alder wood. This only 
warped 2| millimetres, showing a superiority even to 
the old mahogan3\ 

These experiments had been made in each case by 
saturating the under surface of the wood by moisture. 
It seemed, therefore, interesting to see what would 
happen when the conditions were reversed, by putting 
dry calcium chloride inside the box so as to expose 



the under surface of the panel to dry air. When this 
experiment was tried on the three-ply panel it warped 
5-5 millimetres, showing that with the balance of mois- 
ture M'hich it contained, it was more sensitive to drying 
than it was to wetting. 

The next question which I proceeded to investigate 
was the relation between the amount of warping and 
the thickness of the panel. For this purpose I had two 
other panels cut from the same piece of old mahogany, 
one 6 millimetres and the other 12 millimetres thick. 
The amount of warping was 3-5 millimetres and 2-5 
millimetres, so that if we arrange these in order we 
have for the 3 millimetres thick a warp of 5-5 ; for the 
6 millimetres thick a warp of 3-5, and for the 12 milh- 
metres thick a warp of 2-5. Assuming that a piece of 
wood is thoroughly seasoned it is evident that there is 
much less curvature in a thick panel, which confirms 
the mediseval practice of using a panel for even small 
pictures at least 1 in. thick. 

Finally, to get a good comparison between ply-wood 
and old mahogany I tested a five-ply birchwood panel 
from the Venesta Company of 12 milh metres thickness. 
The warp in this case was -5 millimetres. This is the 
ideal panel for artists' purposes. 

The next question investigated was the protective 
value of different treatments in order to reduce warping, 
and for this purpose the white wood panels were used. 
Saturation of a panel on both sides Math a weak solution 
of resin caused very little improvement. A panel sized 
and then coated on each side with one coat of Ripolene 
enamel, only warped 5 millimetres instead of 12 milli- 
metres, showing the protective value of a coat of oil 
or varnish paint. 

An interesting question here arises as to how far a 
coat of varnish or paint on one side is beneficial, the 



other side being exposed to moisture. It is quite possible 
that coating on one side is beneficial as it ■will tend to 
cause the wood to get evenly saturated "svith moisture 
througliout. Another of the white wood panels was 
therefore treated with enamel paint on one side only, 
and the other unprotected side exposed to moisture, 
with the result that it warped 8 millimetres instead of 
12 millimetres. 

From these experiments we can draw the following 
conclusions : — 

That even for a small picture a panel should not be 
less than | in. thick ; the mediaeval panel for a small 
picture being often over 1 in. in thickness. That either 
a good quality of five-ply birch or an old well-seasoned 
mahogany is best. That the panel should be sized and 
painted on both sides with oil or varnish paint." 

The advantage shown by these tests of painting both 
sides of the panel at once raised the question of the 
use of such a panel for tempera painting or oil painting 
on a gesso ground. I found for this problem the following 
solution : — 

A panel was sized and coated with one coat of Ripolene 
enamel. When dry the surface of the Ripolene was 
sandpapered and unprimed canvas attached to the 
surface, in one case with a strong carpenter's glue and 
in the other ease with casein. In both cases the canvas 
was found to be firmly attached, and on being forcibly 
torn off left the glue and the casein attached to the 
Ripolene surface. Such a canvas so attached gives 
a capital surface for a gesso priming of either size or 
casein. This combination of wood and canvas is the 
most reliable backing for a picture which we can obtain 
out of kno\\Ti materials. 



When a large panel is required the planks should 
be tongued and grooved and cemented together Avith 
good glue or casein. Panels are much strengthened by 
the process, if cradling strips of wood, slotted out at 
intervals, are glued on the back the way of the fibre, 
and cross-pieces slipped into the slots. 



The Priming of the Panel 

The best account is that given by Cennino Cennini in 
his Treatise on Painting. He begins by telling us how 
to prepare glue. For the cementing of the portions of 
a panel together he recommends a mixture of cheese, 
well washed with water, and Ume, and he describes a 
fish glue corresponding to our modern isinglass, and a 
glue, " colla di spicchi," which is made from sinews and 
cUppings of goat skins, boiled Avith an equal quantity 
of water till reduced to one-half ; allowed to soUdify, 
cut into shces and put in the wind to dry. He also 
describes a glue made from sha\dngs of the skins of 
sheep and goats, which are softened in water for twenty- 
four hours and then boiled down till the water is reduced 
to one-third, which can be used in place of " colla 
di spicchi " for mixing with the grounds of the picture ; 
and finally a glue made from parchment chppings, 
boiled till the water is reduced to one-third, which he 
uses as a tempera for pigments. He does not consider 
this glue strong enough for the gesso with which the 
panel is to be covered. All these materials and direc- 
tions correspond very closely ^^dth modern methods of 
glue manufacture. 

He fills in hollows and knot-holes with glue and saw- 
dust. He first treats the panel with a thin wash of 
parchment glue and then two coats of strong glue. I 



have often found on examining old panels a very obvious 
layer of glue under the gesso. He next covers the 
panel with strips of old linen soaked in glue. I have 
only very occasionally found old panels prepared in this 
way with linen, so it evidently was not a universal 
practice and was not always followed by Cennino 
Cennini, because in the next sentence he tells us when 
the panel is dry to rasp it with the point of a knife. 
His meaning is somewhat obscure, as he tells us to make 
it quite even. It is, I think, obvious that a slightly 
roughened or rasped surface will make a better tooth for 
the gesso. 

I found on an old Egyptian coffin lid that the fibre of 
the wood had been torn up, and then covered with a 
layer of glue and sand on which the gesso was laid. 
Obviously an excellent practice. 

The gesso used by Cennino Cennini is made from 
plaster of Paris, and is used in two conditions. For 
the first layer he uses plaster of Paris ground with glue 
vrater and laid on AAith the spatula. After two or three 
days to dry it is rubbed do\Mi with a rasp. This is 
called " Gesso-Grosso." 

For the final coat of gesso, he mixes plaster of Paris 
vrith. an excess of water, stirring repeatedly, and keeps 
it wet for a month before use. This slaked plaster of 
Paris he calls " Gesso-Sottili." The first coat ha\*ing 
been rasped and polished, the dried cakes of Gesso- 
Sottili are allowed to soak up water, then ground on 
a muller, mixed mth glue and laid on the panel vrith 
a brush, and rubbed with the fingers into the first 
gesso, giving eight coats, and allo%nng it to dry a little 
between each coat. The whole must dry for two or 
tliree days, and is then rubbed and polished with iron 
rods till it " looks like ivory." 

I have omitted details for the making of raised 



surfaces by moulding. For these the reader had better 
consult Lady Ilcrringham's translation. 

It is e\'ident that a glue corresponding closely to our 
modern carpenter's glue is used, but the amount of 
glue added cannot be estimated, and it is therefore 
not clear whether the finished surface is absorbent or 

From one of his remarks Cennino Cennini evidently 
does not consider the use of Gcsso-Grosso essential. 
The whole can be executed in Gesso-Sottili. Finally, 
I have in many old panels found that whiting has re- 
placed plaster of Paris, and that in some cases oil or 
varnish has been mixed •\\'ith the gesso. The necessity 
for a non -absorbent panel in the case of oil paintings is 
obvious, as the oil would deeply stain the gesso and 
destroy its brilliancy ; and in the case of oil Cennino 
Cennini gives definite instructions to that effect. It 
does not seem to be so essential in the case of yolk of 
egg, but Lady Herringham found it advisable to wash 
over the gesso with a thin egg water. 

Some time ago I obtained possession of a panel which 
had been covered \nth comparatively modern painting, 
and which on cleaning off the modern paint was found 
to be an unfinished oil picture of the Holy Family, 
judging by the style, of the sixteenth century. In this 
case the gesso had been made non-absorbent, but a 
very thin coat of absorbent gesso had been laid over 
it to act as a key for the oil. The lower gesso was com- 
posed of whiting laid over ^^'ith a very thin absorbent 
coat of plaster of Paris. 

Parchment glue from parchment clippings is easily 
prepared and is still used by some gilders ; but excellent 
glues are now manufactured for the use of carpenters, 
and a very fine colourless and pure glue can be 
obtained by taking the gelatine sheets which arc pre- 



pared for bacteriological purposes and dissolving them 
in water. 

Where gilding is required, the gesso is coated with a 
thin layer of bole and the gold leaf cemented on with 
the white of egg, the surface having first been polished. 
The white of egg is thinned with water, and the panel is 
wetted and three or four coats of bole as thin as water 
are laid on with a soft brush, and the bole burnished 
with a coarse linen rag. 

It is of interest to compare this account with the 
practice of Mr. Duncan, R.S.A., a tempera painter of 
to-day. Well-seasoned mahogany is used for the 
panel, and the refined gelatine used for micro-cultures 
is used for the glue. Eight sheets of this gelatine are 
dissolved in a pint of warm water as quickly as possible, 
care being taken not to boil the solution. The panel 
is wetted and a piece of mushn glued on to it. The 
Gesso-Grosso is prepared by mixing plaster of Paris 
with this glue to a consistency which can be plastered 
over the panel with a trowel. This is given two days 
to dry and is then scraped down with the edge of a 
three-cornered file. The Gesso-Sottili is prepared by 
mixing thoroughly one pound of plaster of Paris with 
one gallon of water. This is frequently stirred at first 
and then kept for a month. The water is then drained 
off and the gesso collected in a cloth, the water squeezed 
out, and then mixed with the glue in about equal 
volumes. Eight thin coats of this are laid on with a 
brush and the panel left to dry for four days. The 
surface is then rubbed down with fine sandpaper, and 
in some cases it is burnished with an agate — a tedious 
but excellent practice, as the more carefully the gesso 
is prepared, and the more it is polished, the easier and 
more successful the subsequent painting. It is then 
coated with two or three coats of weak size to make it 



5 i'i^ 




non-absorbent. The back of the panel is also roughly 
coated with gesso. 

In the recipes given by Cennino Cennini, it will be 
noted that there is no mention of any other substance 
than plaster of Paris. As an actual fact, the examina- 
tion of old panels shows that whitening was frequently 
used instead of gypsum. 

If the gesso is to be applied to canvas, he advises 
adding a little starch or a little sugar (Chapter 162), 
doubtless to make it more flexible ; and in his recipes 
for oil painting (Chapter 94), he says, " and in the 
same manner you may work on iron, or on any panel, 
always sizing first " ; and in his recipes in Chapter 90, 
for painting in oil on walls, he states that the surface 
is first to be sized or treated with the whole egg mixed 
with the juice of the fig tree. The surfaces, therefore, 
were probably non-absorbent. 

The early oil pictures, like the tempera pictures, were 
painted on panels carefully coated with pure white 
gesso. This gesso was evidently prepared so as to be 
either non-absorbent or with a very thin absorbent 
layer on a non-absorbent surface, and on this the drawing 
of the picture was made. This is clearly sho^vn in the 
unfinished Van Eyck in the Antwerp Gallery. In this 
picture the blue sky has been painted in, but the rest 
consists of a very fine drawing on the white gesso. Over 
this, according to Van Mander, a thin flesh-coloured 
priming through which the drawing could be seen was 
passed. This priming was of oil. 


To begin at the beginning, the canvas itself should 
be woven of pure flax fibre, unbleached or very slightly 
bleached, so as to secure the greatest possible durability 

E 65 


of material. The test of time has shown that a linen 
canvas is very durable, pictures painted on canvas of 
the fifteenth century being in existence to-day. They 
have had to be relined, that is, backed by another 
canvas in most instances ; but this has been due to 
defects in the priming, resulting in the crumbling and 
separating of the paint. Linen canvas of unbleached 
and partly bleached flax can be obtained from the 
artists' colourman. 

Canvas has two great defects. In the first place it 
does not protect the back of the paint film from the 
attack of air and moisture. In the second place it is 
always changing, owing to the fibres readily absorbing 
moisture from the air. This movement is caused either 
by change of temperature, which alters the saturation 
point, or by change in the amount of moisture in the 
air. This constant movement tends to crack the paint 
and to loosen the priming from the canvas, and is 
probably the reason why relining is so often necessary. 
It also vibrates mechanically. 

Having obtained a suitable canvas, the next problem 
is how to prime it. There can be no doubt that the 
earlier canvases in the fifteenth and sixteenth century 
were primed with gesso, the gesso priming being 
gradually replaced by an oil priming. The accounts 
given by Vasari and by later ^vl'iters of how to prime 
in oil are too vague to be of much practical use. When 
priming with gesso, Cennino Cennini advises intro- 
ducing a little honey to make the canvas more flexible 
— a very doubtful practice. The objection to gesso 
priming is that it is apt to crack off when the canvas 
is rolled up, and is easily caused to peel off by exposure 
of the back of the canvas to damp. A very thin priming 
is safest, as it is less apt to crack off. 

If picture restorers are to be believed, such gesso 



primings were long used after the introduction of canvas, 
and were usually treated with a sufficient proportion of 
size to make them non-absorbent. 

Vasari gives an account of how to prepare both 
panel and canvas either with a gesso or an oil priming 
(Vasari on Technique, translated by Louisa Macle- 
hose) : — 

How TO Prime the Panel or Canvas 

" I must now explain how to set about the work. 
When the artist wishes to begin, that is, after he has 
laid the gesso on the panels or framed canvases, and 
smoothed it, he spreads over this with a sponge four 
or five coats of the smoothest size, and proceeds to grind 
the colours with walnut or linseed oil, though walnut oil 
is better, because it yellows less with time. When they 
are ground with these oils, which is their tempera 
(medium), nothing else is needed, so far as the colours 
are concerned, but to lay them on with a brush. But 
first there must be made a composition of pigments 
which possess siccative qualities, as white lead, driers, 
and earth, such as is used for bells, all thoroughly well 
mixed together and of one tint, and when the size is 
dry this must be plastered over the panel and then 
beaten with the palm of the hand, so that it becomes 
evenly united and spread all over, and this many call 
the ' imprimatura ' (priming)." 

This account is very difficult to follow. In any 
pictures I have examined, the paint lies directly on 
the white gesso. There is some doubt as to the nature 
of the " earth for bells." 

Then follows another recipe for preparing canvas 
with oil pruning which is as follows : — 



Painting on Canvas 

" In order to be able to convey pictures from one 
place to another, men have invented the convenient 
method of painting on canvas, which is of little weight, 
and when rolled up is easy to transport. Unless these 
canvases intended for oil painting are to remain 
stationary, they are not covered with gesso, which 
would interfere with their flexibility, seeing that the 
gesso would crack if they were rolled up. A paste, 
however, is made of flour and of walnut oil, with two 
or three measures of white lead put into it, and after 
the canvas has been covered from one side to the other 
with three or four coats of smooth size, this paste is 
spread on by means of a knife, and all the holes come 
to be filled up by the hands of the artist. That done, 
he gives it one or two more coats of soft" size, and then 
the composition of priming. In order to paint on it 
afterwards, he follows the same method as has been 
described above for the other processes." 

" The composition of priming " is probably the one 
given in the former recipe. 

Giovanni Battista Armenini was born in Faenza 
about 1530, and was trained as a painter, but ultimately 
became a monk. In 1586 his treatise, De Veri Precetti 
della Pittura, was published. He states that before 
sizing, the holes in the canvas can be filled up with 
flour 'paste and one-third white lead, and that the 
canvas should be sized both back and front. This is 
apparently, therefore, different from Vasari's state- 
ment, though it looks suspiciously like the same recipe 
incorrectly described either by the one author or the 
other. The priming is to be composed of white lead, 
with giallolino and terre de campane, or with verdigris 

68 ' 


and umber. The additions of verdigris and umber are 
as driers, and the amount added may have been very- 
small so as merely to tint the white. He himself advises 
a priming of a light flesh or flame colour, which he 
brings about by the addition of varnish. He states 
that pictures with dark grounds ultimately darken, and 
that the oil in the ground darkens and sullies the colours 
and therefore those who wished to prevent change made 
their grounds of white lead with one-sixth part varnish 
and a little red. 

With these recipes may be compared some given by 
De Mayerne : — 

* " After spreading your canvas on a frame, give it 
some gliie of scrap of leather or size. . . . When the 
glue is dry, prime quite lightly with brown-red or dark 
Ei_nglish red. Leave to dry ; smooth with pumice stone. 
Then prime with a second and last layer of white lead, 
carefully chosen cliarcoal, small coals, and a little umber, 
that it may dry more quickly. A third layer may be 
given, but two are sufficient." 

This would doubtless produce a grey priming. It is 
evident from the introduction of moulding clay by 
Vasari, and the use of a red-ochre clay by Wallon, w^ho 
supplied this recipe, that there was some fear of a pure 
white lead priming, at any rate in direct contact with 
the size. 

Abraham Latombe, of Amsterdam, supplies this 
recipe to De Mayerne : — 

" Canvases must first be glued with calf or goat- 
skin glue ; the whole artifice consists in this. For if 
the glue is too strong, the canvas easily splits and tears. 
After putting the glue on the canvas, lay it while still 
damp on the marble, flatten with the rubber all joints 



and knots ; then let it dry. Then prime with white 
lead and a little umber. One priming is sufficient, but 
if you give two layers the canvas will be more even. In 
painting landscapes let your priming be very light in 
colour. A siccative oil prepared with litharge should 
be used." 

De Mayerne comments on this, that he found the 
colour separated from the canvas in a picture by this 
artist which had been hanging on a damp wall. He 
therefore objects to size and also to umber. In another 
part of the manuscript, which contains the recipe of 
" the little painter of M. de St. Jehan "— 

" after sizing we are directed to prime with bole, | lb., 
and umber, 2 oz., ground in oil. Smooth with pumice 
and finally prime with white lead, 1 lb., and umber, 
1 oz." 

Similar recipes are given by other -^Titers. 

It is evident from all these recipes that the canvas 
was first prepared with size ; that sometimes the holes 
were filled up with flour paste and white lead, or flour, 
white lead, and oil (some obscurity here) ; that the 
priming was white lead with a drier like umber, and 
sometimes varnish, either mixed with a clay-like sub- 
stance (Vasari, Armenini) or underlaid with a priming 
of a clay-like substance (bole, De Mayerne) ; that the 
priming was white, light grey, or brightly tinted, of a 
flame colour ; that Armenini regarded the white as 
of great importance. It also seems that the white lead 
and oil alone in direct contact with the sized canvas 
was not considered safe. Some modern recipes consist 
of mixtures of white lead and china clay, probably a 
very sound practice. 



This, the most famous of Holman Hunt's pictures, is an interesting ex- 
ample of the special technique employed by the Pre-Raphaelite Brother- 
hood which has proved to be somid, by the excellent condition of their 
pictures. On a hard, well-matured white priming, white lead stiffly mixed 
with copal varnish was laid, and thin translucent painting in oil laid on 
the still sticky priming. The absence of cracking, and the well-sustained 
high colour-key. is to be expected from this method of painting in the light 
of the most recent scientific researches into the properties of oil and pigments. 
For further information the reader is referred to Chapters XI, XII. 


It is evident that no very clear and definite informa- 
tion can be obtained from these recipes, and I know of 
no others that are any better. 

In conclusion, the following account by Mr. Holman 
Hunt of the method followed by the Pre-Raphaelite 
Brotherhood is interesting : — 

" Select a prepared ground originally for its bright- 
ness, and renovate if necessary with fresh white when 
first it comes into the studio ; white to be mixed with 
a very little amber or copal varnish. Let this last coat 
become of thoroughly stone-like hardness. Upon this 
surface complete with exactness the outline of the part 
in hand. On the morning for the painting, with fresh 
^/hite from which all superfluous oil has been extracted 
by means of absorbent paper, and to which again a 
small drop of varnish has been added, spread a further 
coat very tenderly with a palette-knife over the part of 
the face work, and of such a consistency that the drawing 
should faintly shine through. In some cases thickened 
white may be applied to the pieces needing brilliancy, 
with a brush and the aid of rectified spirits. Over this 
wet ground the colours, transparent and semi-trans- 
parent, should be laid with light sable brushes, and the 
touches should be made so tenderly that the ground 
below shall not be worked up, yet so far enticed to blend 
with the superimposed tints as to correct the qualities 
of thinness and staininess which over a dry ground, 
transparent colours used, would inevitably exhibit. 

" Painting of this kind cannot be retouched except 
with an entire loss of luminosity." 

Before coming on to consider further the question of 
priming, Mr. Holman Hunt's account of the method of 
painting of the Pre-Raphaelite Brotherhood is worthy 
of some discussion. 



In the first place it will be noted that these pictures 
are painted on a white lead oil or rather white lead oil 
copal priming, in which the amount of medium was 
reduced as much as possible, so as to have as little 
yellowing of the ground in the course of time as possible, 
and evidently plenty of time was allowed for the oil in 
the first priming to get thoroughly oxidized, for he says 
keep it until it is stone hard. On this, before painting, 
another very thin coat of white lead with a minimum 
of oil and copal is laid, and transparent pigments are 
glazed on to this wet surface. 

We have here, then, in so far as the use of a white 
ground and transparent pigments are concerned, the 
sixteenth-century oil technique, while, as we shall find 
when we study the properties of drying oils, the painting 
on to the sticky and still wet or partially wet paint is 
a perfectly sound and safe thing to do, the danger of 
cracking coming later when the oil has surface dried. 

How the Pre-Raphaelite Brotherhood evolved this 
particular technique I do not know ; but it will be 
evident from what we will learn, both about the optical 
properties of oil and pigments and the properties of oil 
in drying, that they followed a sound method which 
accounts for the excellent condition and brilliant 
luminosity of their pictures. 

If we wish to paint in tempera or in a tempera oil tech- 
nique we require a gesso ground. A gesso ground has 
not got the elasticity of an oil ground, and consequently 
is more affected by the expansion and contraction of the 
canvas with changes in the moisture of the air, and by 
the vibration of the canvas, so that even if protected 
from damp it is doubtful if it will stand the test of time. 
Cennino Cennini's suggestion of adding honey is a 
doubtful expedient, as that means that a gesso will 
always be slightly moist. If he had suggested a certain 



amount of lime instead of gypsum with the honey, 
there might have been something in it, as the use of 
sugar mixed with Hme to produce a very hard and 
durable mortar is well known in India. Size is not the 
only material which can be used to make a gesso 
ground, and many painters to-day prefer casein. Mr. 
Batten has used it now for a long time to prepare his 
grounds, and has found it reliable. 

The following is the account given by Mr. Batten 
in the Transactions of the Tempera Society for 1901- 
1907, of how to prepare a tempera ground with 
casein ; — 

" To Prime the Canvas. — Choose an oil-primed canvas 
of good texture. Pin it face downwards on a board or 
on the floor. Use plenty of drawing pins and push them 
well in. Allow a margin on all sides of 1| or 2 in. beyond 
the ' sight ' measure of the picture. See that the back 
of the canvas is clean and free from dust. Rub it lightly 
with a benzined rag to remove any superficial grease. 
To prime 8 sq. ft. of canvas, put one ounce of casein^ 
into eight ounces of water, and stir for a quarter of an 
hour with a wood or bone spoon. Then add strong 
ammonia drop by drop — stirring all the while — till the 
casein dissolves in the water and forms a strong size. 
Keep on stirring till the size has the consistence of a 
thick honey and flows from the spoon without clots. 

* Samples of casein differ very much in strength, so the proportion 
of one of casein to eight of water cannot always be reUed on. It is 
well to test each fresh purchase of casein so as to ascertain the right 

I have tried using lime in place of ammonia in making casein eize, 
but I did not find any advantage in so doing. 

For Cheese glue see Cennino, chap. 112, and Theophilus, Book I, 
chap. 17. 

A method of priming with casein size and zinc white is described 
in Vibert's Science de la Peinture. If zinc white is used it is necessary 
to add a little glycerine to the size. 



The size ought not to be dikited with water after it is 

" Take plaster of Paris that has been slaked in water 
for more than three weeks and squeeze it tightly in a 
linen cloth. Of this take a lump nearly equal in bulk 
to the casein size. Mix it thoroughly with the size, and 
work it with a spoon through a hair sieve. Take a large 
hog's hair brush and lay this priming on the canvas, 
working it well in. If a rough surface is wanted a second 
coat may be dabbed on before the first is quite dry. If 
a smooth surface is wanted, let the first coat dry and 
then sweep on another. The priming is better for 
drying fairly quickly, so if the weather is damp a good 
fire should be kept in. Undue inequalities of the surface 
may be scraped down with glass paper, or with a knife, 
after the priming is perfectly dry. 

"To make a backing for the canvas.— Take one or 
more sheets of ' wire-wove roofing ' ^ cut to the size of 
the picture and fasten on to a wooden frame or stretcher 
by means of brass screws and copper nails. Mix equal 
quantities of Venice turpentine and oil-copal varnish 
(commercial). Add white lead in powder and grind 
into a putty of such consistence that it can be spread 
with a palette-knife, but is too thick to lay on with a 
brush. With a palette-knife spread a fairly thick coat 
of putty over the surface of the ' wire-wove roofing ' 
as evenly as possible. Roll up the prepared canvas 
with the oil-primed surface outwards. Do not cut off 
the margin yet. 

" Apply the roll of canvas to one edge of the puttied 
surface and unroll slowly, pressing the canvas do-w-n 
with a photographer's roller squeegee. Work from the 

1 " Wire-wove roofing " is millboard enclosing a mesh of iron wire. 
It can be obtained of the W.W.R. Company. As it is difficult to cut, 
it is well to be careful to order the exact size. 



middle outwards, so as to avoid enclosing bubbles of air. 
If any air should have been enclosed prick the blister 
■with a needle and press again with the squeegee. 

" If after a few days it is found that any part of the 
canvas is not stuck do^vn firmly, it can be pressed down 
with a hot iron. 

" Cut off the margin of the canvas." 

As it is hygroscopic, the exclusion of glycerine, which 
is given in Vibert's recipe, is wise, as also is the exclusion 
of oil which is found in other recipes. The introduction 
of oil in the presence of alkali, whether soda, potash, 
or ammonia, means saponification of the oil, and the 
behaviour of such a blend would require very careful 

I find that Mr. Batten's grounds are very firmly 
attached, but they are not so flexible as an oil ground, 
and here it may be plainly stated that for painting in 
tempera, where some form of gesso is essential, canvas 
without backing and support is unsuitable. 

The preparation of a wood panel has already been 
fully described. 

Canvas can, of course, be glued to a wood panel, 
which is only a revival of the best ancient practice. 

In the case of a picture to be painted throughout in 
oil, there is not the same need for a gesso ground. No 
doubt a gesso ground is preferable to an oil priming 
from the optical point of view, as it will be, and will 
remain, whiter ; but in the case of a tempera picture we 
must have a size or casein preparation, as it must be a 
preparation to which an egg or size film will attach 

In accepting an oil ground for a picture painted in 
oil throughout, we at once accept a lower colour key 
from the beginning, and one which will become still 



lower with time. On the other hand, we have the 
advantage of the Hght, easily handled canvas on its 
stretcher. The traditional preparation, sizing and then 
priming with oil and white lead or with a mixture of 
white lead with other pigments, seems perfectly sound. 

The attainment of a firmly adhering coat depends on 
three things. 

The size must be sufficient to protect the fibre from 
the oil, but not more than sufficient. It is a good plan 
to roughen up the surface. The white lead should be 
a good stack lead, and the proportion between oil and 
white lead requires careful adjustment so as to ensure 
a firmly adhering and sufficiently elastic surface which 
at the same time contains a minimum of oil. The paint 
must be well stubbed and worked into the canvas, a 
little at a time, with two or three weeks between each 
treatment to allow it to dry. Some prefer to use pure 
white lead, others, to begin with, white lead and china 

An excellent mixture is white lead containing some 
10 per cent of barytes. This would be less likely to 
crack, and will give a tooth to the canvas. 

The next question is the selection of the oil. If we 
take a sample book of primed canvases, and cutting off 
one-half of each expose them to light while placing the 
remainder in the dark, and we then compare the two 
halves after six months, it will be found that the amount 
of yellowing differs very much on different canvases, 
showing very different methods of preparation or very 
different oils used by the same firm. The cause of these 
differences is well worthy of the investigation of artists' 

No doubt, if poppy or walnut oil has been used, a 
priming would be obtained which will yellow less than 
linseed oil, but in the light of Professor Eibner's re- 



searches it is obviously safer to use linseed oil for 

Of linseed oil the best is Dutch Stand oil. This has 
been clearly brought out both by Professor Eibner's 
researches and by my own experiments on the darkening 
of oil films. There are obvious objections to using it to 
paint with, but it could be, and should be, used for 
priming. It is too sticky to grind sufficiently stiff with 
the pigment alone, and requires thinning with turpen- 

The amount of thinning with turpentine requires to 
be tested by experiment, so as to obtain a sufficient 
bind with a minimum of oil. The amount of oil required 
to bind a pigment is much less than the amount required 
to grind a pigment, especially with thick and sticky oils. 
For example, from experiments with boiled oil and 
lithophone I found that 5 grammes of lithophone was 
bound by -5 grammes of oil, but the same quantity of 
lithophone required 1 gramme of oil to grind it. 

The proper dilution with turps having been found, 
the canvas can then safely be primed with Stand oil 
and stack white lead. The following proportions will 
not be found very far from correct : 20 parts of white 
lead, 2 parts of barytes, 1 part of Stand oil, and sufficient 
turpentine to make quite a thin paint for the first appli- 
cation, the turpentine being reduced in future coats so 
as to give the paint a stiff consistency when applied. 
A sound copal varnish would do excellently well, or 
probably still better a Stand oil in which a resin had 
been dissolved. 

The canvas having been primed on the face should 
now be turned over, and the back treated with a wash 
of tannic acid or formalin to make the size insoluble. 

The experiments described elsewhere on the yellowing 
of the oil have shown the importance of protecting the 



canvas from the back. Professor Oswald has suggested 
an excellent method of doing this. The canvas is covered 
with tin-foil cemented on with gold size. Tin-foil is 
practically indestructible, and has been used from the 
earliest times in the decorative arts, and will make the 
back of the canvas impervious to moisture. 

The other method is to coat it with beeswax and 
resin. Beeswax and resin is the oldest varnish of which 
we have a record, having been used by the Greeks to 
varnish their ships. It is remarkably resistant to the 
passage of moisture or injurious gases. If we dissolve 
some size in water, colour it with neutral htmus and 
paint it out on sheets of glass, and when dry, varnish 
with different materials, and then place the varnished 
glass sheets under a bell jar with a strong solution of 
hydrochloric acid, the moist hydrochloric acid gas, when 
it has succeeded in penetrating the protective coating, 
turns the litmus red. If the surface is painted with 
beeswax in turps and then polished, the litmus turns 
red in four or five minutes. If coated with mastic 
varnish it resists some three or four hours. A layer of 
beeswax and resin, 3 to 1, less than tV iri- thick resists 
for several days. Beeswax is slowly oxidized in time, 
and is not so resistant to chemical change as paraffin 
wax, but it has the advantage of being sticky and 
yielding to expansion and contraction without cracking. 
To protect with beeswax and resin melt them together 
and lay on the melted mixture freely on the back of 
the canvas with a brush and distribute with a steel 
palette-knife. While it is still sticky, lay on it strips 
of canvas and press them home with the palette-knife. 
Cover the surface with a further supply of the melted 
wax, and distribute over the surface with a steel knife 
as before. This makes an excellent and permanent 
backing to a picture. 



The canvas when primed should be kept six months 
before it is painted on, to enable the oil thoroughly to 
dry, and before used should be rubbed over with fine 

If a painter wishes to use a gesso priming, the safest 
way is as follows : Take an oil-primed canvas ; cover 
the oil-primed surface with tin-foil or resin and canvas, 
and prime the raw canvas surface with a very thin 
priming of prepared plaster of Paris and gelatine or 

If the canvas is purchased ready primed, it should in 
the first place be vigorously worked between the hands 
to see that the priming is properly attached. 

Primed canvases differ very much in the extent to 
which they yellow with time. This can be tested, as 
mentioned on a previous page, by obtaining a booklet 
from the artists' colourman, cutting a little piece off 
each piece of canvas and attaching them to a sheet of 
glass, which is put in a window, and putting away the 
booklet wrapped up in dark paper in a drawer. After 
three months the exposed pieces can be compared with 
the pieces in the booklet, and the priming which changes 
least in colour selected. Keep the primed canvas six 
months before painting on it to ensure the thorough 
drying of the oil priming. 





Whites, Blacks, the Earth Pigments, Vermilion, 
Cadmium, and the Yellows 

I DO not propose to follow any logical order in describing 
the pigments used by the painter, but rather to discuss 
them in the way which seems easiest to bring out their 
properties. The subject is a difficult one, because the 
pigments are derived from so many sources that for 
their complete understanding the whole range of 
chemistry is required. The painter is interested in their 
properties, but not in their origin, and yet some account 
of their origin is necessary in order to understand their 

For fuller details than are given here the reader is 
referred to Chemistry of Paints and Painting, by Sir 
Arthur Church. The most important of the pigments 
are Whites. We shall begin, therefore, by discussing 
the various white pigments available for use. 

White Lead 

Of these the first is white lead, known to the painter 
as flake white, Kremnitz white, and silver white. This 
pigment, known in the time of Pliny, has been used by 
painters from classical times, and the best white lead 
is still made by the process described by Pliny, namely, 
the exposure of sheets of lead to the vapour of acetic 



acid, little pots of acid and the grids of lead over them 
being packed round by spent tan in a state of fermenta- 
tion. Under these conditions, by a somewhat complex 
series of reactions, metallic lead is converted into a 
white friable mass which we know as white lead and 
which is probably a combination of lead carbonate and 
lead hydrate. It is separated from the unattacked lead, 
washed and ground and dried and is then ready for the 
market. It is a pigment of excellent body and covering 
power, and enters into a partial saponification with the 
oil, which results in a paint which has proved most 
durable under the test of time. 

There are other methods of preparing it, such as the 
precipitation of a pure carbonate of lead, but these 
preparations have not the body of white lead made by 
the old process, and do not enter into such intimate 
connection with the oil. On the other hand, they have 
the advantage of not yellowing the oil quite so much as 
the lead made by the old process. 

The artists' colourman usually uses Kremnitz white, 
which is a very pure white. It is named after a to"wn in 
Czecho-Slovakia, and is prepared by the action of 
acetic acid vapour and carbonic acid gas on litharge. 

White lead has two defects. In the first place it is 
very poisonous, and while such cases are rare, it has not 
been unknown for artists to get poisoned by white lead. 
Care should be taken, therefore, to remove white lead 
by thorough washing of the hands after handling. It 
is much more poisonous in the form of powder, but the 
painter as a rule only handles it ground in oil. 

The other defect is the susceptibility of white lead 
to sulphuretted hydrogen gas, which turns it black, 
forming sulphide of lead. Fortunately, this compound 
is decomposed on exposure to a bright light, forming 
the white lead sulphate, and also can be bleached by 

F 81 


peroxide of hydrogen. Peroxide of hydrogen can be 
appKed in water sohition to an oil picture and dissolved 
in ether to paintings in gum or size, like some of our old 
water-colour paintings. In modern cities this gas is 
not prevalent, and the conditions of pictures of all ages 
painted in white lead show that in practice there is not 
much danger .from this source in the atmosphere of a 
private house or gallery. It has now been rightly 
excluded from the water-colour palette, being replaced 
by zinc oxide, but still remains the best pigment for 
painting in oil. It is usually described as an amorphous 
opaque pigment, but as an actual fact consists of 
doubly-refracting transparent crystals, as can be seen 
by grinding it in bromonapthalene, and owes its white- 
ness to its high refractive index, just as we find in another 
chapter that powdered glass is apparently an opaque 
white body. Students interested in the microscopic 
examination of pigments are referred to the article in 
Chemical and Metallurgical Engineering, by Henry Green, 
(January 10, 1923), where the crystalline character of 
white lead and zinc white is described. Possibly owing 
to slow saponification, but more probably owing to the 
increase in the refractive index of oil with time, a layer 
of white lead becomes gradually more transparent, the 
particles of white lead in oil in the Rokeby Venus, when 
examined under the microscope, appearing much more 
translucent to light than freshly ground white lead of 
to-day. The importance of this in helping towards the 
lowering of tone of oil pictures will be dealt with later. 
It is essential for artists' use that it shall be quite free 
from lead acetate, which is sometimes present as an 
impurity, owing to the method of manufacture. It not 
only causes the white lead in oil to yellow badly, but 
may cause darkening of mixtures of white lead and 
ultramarine, cadmium yellow, and vermilion. Eibner 



has shown that these pigments ground in oil, when pure 
and mixed with pure white lead, do not change. 

Zinc White 

The next white pigment of importance is the oxide 
of zinc or zinc white, which is known to the water-colour 
painter as Chinese white. It is preferred by some 
artists to white lead, on account of the fact that it is 
not affected by sulphuretted hydrogen, and it is an 
excellent pigment, but does not form so flexible a coating 
in oil as white lead, and is considered by some to be 
more apt to crack, although I have no positive evidence 
of this, and am disposed to think that the cracking is 
due to the use of poppy oil to grind the pigment. It 
has the defect that it requires a much larger proportion 
of oil to grind the pigment than white lead, and conse- 
quently while not changing in colour itself, is more 
likely to go down in tone, owing to the yellowing of the 
oil, than white lead. The proportion of oil required to 
grind white lead as compared with zinc white being as 
follows : for white lead, 100 parts take 15 parts of oil, 
and zinc white, 23 parts of oil, accordmg to figures 
supplied by Messrs. Winsor and Ne^sHon to Sir Arthur 
Church. The amount varies a little according to the 
practice of the grinder, but may be taken roughly to be 
at least double the amount required for white lead. 
Ground in bromonaphthalene it is seen to consist of 
transparent particles. 

Lead Sulphate 

Lead sulphate and zinc oxide and barium sulphate 
are among the other permanent whites supplied by the 
artists' colourmen. A new white, titanium white, has 
recently been introduced. More experience in the 



behaviour of this pigment is required before introducing 
it into the artists' palette. 


Lamp Black, Chaecoal Black, Ivory Black. All 
the blacks usually used by artists owe their pigmentary 
value to carbon, though other blacks are known. Carbon 
is the residue left on raising to a red heat various organic 
substances. We have, for instance, charcoal black, 
best prepared from vine clippings ; lamp black, from 
the burning of oils and fats without sufficient air ; and 
ivory and bone black, from heating ivory and bone 
fragments. Charcoal and lamp black are almost pure 
carbon. Ivory and bone black contain as well as carbon, 
phosphate of lime of which the bone is composed. 
They are all permanent, and all that is necessary for 
both the ivory black and lamp black is that they should 
be quite free from greasy and tarry material which will 
prevent their drying properly and also is apt to cause 
cracking. I have so frequently seen fine cracks appear 
in lamp black when painted in oil over white, that in 
my opinion ivory black is preferable. Whether this 
cracking is due to the physical properties of lamp black, 
or to the presence of greasy material which could be 
removed by washing with petroleum before grinding, is 
worthy of investigation. 

Very dense and pure blacks are now obtainable from 
the burning of natural gas, obtained in the oil regions in 
America, with a limited supply of air. They are used 
in the paint trade and are the finest blacks on the market, 
but have not yet, I believe, been introduced by the 
artists' colourmen. 

Graphite, another variety of carbon, is also obtain- 
able as a pigment, and is absolutely permanent. 



The Earth Pigments 

Yellow and Red Ochre, Raw and Burnt Sienna, 
Venetian and Indian Red, Terre Verte, Raw and 
Burnt Umber. Having dealt briefly with the whites 
and the blacks, the next group of the greatest interest 
to the painter is the red and yellow ochres, terre verte, 
and the siennas. These pigments, which are obtained 
by mining, grinding, washing and floating, may be 
roughly described as clay stained with compounds of 
iron. The hydrated oxide of iron is yellow, or some- 
times, in the case of certain compounds, green, as we 
find in terre verte ; while, if these yellow bodies are 
heated, we get the red ochres which owe their colour 
to the anhydrous oxide of iron. 

The yellow ochres are quite permanent, and can be 
safely mixed with other pigments, though the choice 
of the right yellow ochre for use in Buon-Fresco, as is 
stated in the chapter on that subject, is a little open to 
doubt. It will be seen from the table in Chapter IX 
that French yellow ochre is one of the least transparent 
of pigments, and therefore will not be very much 
affected by the optical changes in linseed oil. Ground 
in methylene-iodide saturated with sulphur, it is seen 
to consist of translucent isotropic particles. The 
red ochres, light red, Venetian red, and Indian red are 
either found in nature or are produced by roasting the 
yellow ochre, or by heating in a retort certain com- 
pounds of iron, like green vitriol (iron sulphate), in 
which case they are practically pure oxide of iron. 
Care should be taken in the case of the artificially pre- 
pared oxides that they are free from partially 
decomposed sulphate of iron. All these red oxides are 
absolutely permanent, and are the only pigments, as 



can be seen in the table in the chapter on Oil Painting, 
of those tested which have proved to be opaque. 

The Mars yellows and reds are artificially prepared 
yellow and red oxides of iron compounds, and are also 

Sienna. Raw sienna also owes its colour to the 
presence of compounds of iron and maganese, but is 
more transparent than the ochres, coming between 
pigments like madder lake and chrome yellow in trans- 
parency. It is a permanent pigment, but owing to its 
transparency and the rise in refractive index of linseed 
oil, gets deeper in tint in course of time, which accounts 
for the suspicion with which it has been regarded by 
artists. On roasting this pigment we obtain burnt 
sienna, which is also a permanent pigment. 

Umbers. Umber owes its colour to the presence not 
only of compounds of iron, but also of compounds of 
manganese. There are two varieties : raw umber, as 
it is obtained from the mine ; and burnt umber, which 
is produced by roasting raw umber. These pigments 
are quite permanent. 

There are other browns to be found in the list of 
artist's colours, such as Vandyke brown. Unfortunately 
Vandyke brown is a name applied to pigments from 
different sources, some of which are fugitive to light. 
A Vandyke brown, therefore, should be carefully tested 
by exposure to light before being used. Owing to the 
presence of manganese, which is a powerful drying agent 
for linseed oil, pigments ground in umber will be found 
to dry faster than pigments in other media. 

We have now dealt with the earth colours, and shall 
go on to consider the more important of the artificial 
colours in their order in the spectrum. 

According to Eibner the yellow ochres, siennas, and 
umbers dissolve slightly in linseed oil, forming dark 



coloured liquids which cause lowering of tone and stain 

Vermilion. This is one of the oldest pigments in the 
artist's palette. Originally the native sulphide of 
mercury or cinnabar was used, but even in the time of 
Pliny the preparation of vermilion by the subliming 
together of sulphur and mercury was kno^vn. This is 
the method used to-day in China, but has been replaced 
in Europe by wet methods of manufacture. Vermilion 
under ordinary conditions in the light of a room or 
gallery is permanent, as can be seen by examining 
some of the old pictures in which vermilion has been 
used, such as " The Rape of Helen," National Gallery, 
No. 591. When exposed to sunlight the sulphide of 
mercury is converted from the red to the black variety, 
thus going brown. This is a molecular change and 
cannot be prevented by protection with varnishes. 
It is somewhat capricious, some samples standing 
exposure without change, but it is evident that vermilion 
cannot be safely used for outside painting. Sometimes 
the vermilions made in this country contain residue 
of alkaline sulphide, in which case they are not safe to 
mix with white lead. The Chinese vermilion is almost 
chemically pure, containing less than one part in a 
thousand of a harmless neutral ash. Pure vermilion can 
be safely mixed with pure white lead without any 
change. The opacity of vermilion is about the same as 
chrome yellow, and it is almost to be regarded as an 
opaque pigment ; its place in the table will be found in 
the chapter on Oil Painting. 

Red lead is an oxide of lead obtained by roasting 
massicot gently. It is occasionally found in mediaeval 
manuscripts, and when it has not turned black, has stood 
well. Modern tests have shown that it is not permanent 
either in oil or water colour. 



Cadmium Scarlet. A comparatively new pigment 
in the artist's palette is cadmium scarlet. The cadmium 
yellows are compounds of the metal cadmium with 
sulphur. Closely related to sulphur is another element 
known as selenium. The cadmium scarlets are com- 
pounds containing both sulphur and selenium. Tests of 
exposure to light show them to be remarkably perma- 
nent, and they form a valuable addition to the artist's 

Oranges and Yellows. Two oranges are available, 
cadmium orange and chrome orange. They are both of 
doubtful stability. The cadmium orange on exposure 
to light changes into cadmium mid, and the question of 
the permanency of chrome orange will be discussed 
when we come to chrome yellow. The cadmium yellows, 
including cadmium orange, are sulphides of the metal 
cadmium. By varying the methods of manufacture a 
series of tints can be obtained, from a very pale yellow 
up to cadmium orange. Of these the only one suffi- 
ciently permanent for oil painting is what is sometimes 
called cadmium yellow, and sometimes cadmium mid, 
and which can be obtained by the precipitation of 
a cadmium salt in acid solution with sulphuretted 
hydrogen gas. The paler varieties either fade completely 
when exposed to light or turn a deeper tint. There is 
only one way in which a pale cadmium can be made 
which is reasonably permanent. If cadmium sulphide 
is precipitated with sulphide of zinc, or hydrate of 
zinc, or of magnesia, and is then heated to a temperature 
of from 500° to 600° C. for a couple of hours, a beautiful 
pale yellow is obtained which resists exposure to light. 
This is the only pale cadmium which should be used by 
the artist. 

It is also possible to make a permanent orange cadmium 
in the same way in which cadmium scarlet is made, but 



with less selenium. As far as I am aware, no such orange 
cadmium is on the market. Ground in methylene-iodide 
saturated with sulphur, the cadmium yellows are 
translucent bodies. 

Chrome Yellows. The chrome orange and chrome 
yellow are chromates of lead. They are very beautiful 
pigments and on the whole moderately permanent, but 
unfortunately are not above suspicion. If acted upon 
by sulphuretted hydrogen they are decomposed, and 
the colour cannot be restored by exposure to light, as 
in the case of white lead. Moreover, under certain 
conditions they tend to become greenish in hue. I 
have seen chrome yellow on an oil sketch by Constable 
as perfect as the day it was painted, while, on the other 
hand, the chrome yellows in experiments made by Mr. 
Holman Hunt are all greenish in colour. Further 
investigation is required to find out the conditions of 
stability for this pigment. Lead chromate ground in 
bromonaphthalene is seen to consist of transparent 
double refracting crystals. The orange chromes are 
basic chromates obtained by treating the normal chro- 
mate with caustic soda. It is a remarkable fact that in 
spite of modern chemical discoveries, the bright yellows 
remain the weak point in the artist's palette, just as 
they were in earlier times, when the only bright yellow 
available was orpiment, the yellow sulphide of arsenic, 
a most beautiful pigment ; but, if the experience of the 
older painters is to be trusted, dangerous to mix with 
any other pigment. So great were their difficulties in 
the matter of a bright yellow that I have found Holbein 
using gold leaf and yellow ochre to obtain a high point 
of light in yellow — an excellent device. 

Cobalt Yellow or Aureoline. Cobalt yellow is a 
complex salt of potassium and cobalt which, judging 
by its composition, would not be likely to yield a stable 




pigment, but the tests made by Captain Abney and 
Professor Russell showed it to be quite reliable in water 
colour. It is a potassium cobalt nitrite obtained by 
precipitating cobalt sulphate with potassium nitrite in 
the presence of acetic acid. When used in oil it must 
be remembered that it is a very transparent pigment, 
and that if too much oil is used it will go brown. 

Barium Chromate (lemon yellow). There is more 
than one pigment sold under the name of lemon yellow. 
It may be a pale lead chrome, a strontian chromate, 
or barium chromate. If it is barium chromate, ob- 
tained by precipitating a barium salt with potassium 
chromate, it is an absolutely permanent and reliable 

Zinc Chromate. This is a beautiful yellow, but it 
is not permanent in oil, and is slightly soluble in 

Orpiment. This is a sulphide of arsenic, and can be 
obtained naturally or prepared artificially by heating 
sulphur and arsenic together in a covered crucible. 
The orpiment collects on the cool surface of the cover 
by precipitating the sulphide from a solution of arsenic. 

It has been used from early times on illuminated 
manuscripts, and has also been used in oil painting. 
Old writers complain of its slow drying, and state that 
it must not be mixed with white lead or copper car- 
bonate, which is quite likely correct as it is a sulphide. 
It is described as fugitive by Sir Arthur Church and 
other writers. Sir Arthur Church quotes an experiment 
of Sir Joshua Reynolds in support of this, in which 
orpiment in oil has faded, while orpiment in Venice 
turpentine has lasted. It has lasted perfectly in illumin- 
ated manuscript, and it seems unlikely that the mineral 
itself is not permanent. Further experiments on this 
would be of interest. It is a beautiful yellow, but very 



poisonous, and is not very likely to be reintroduced into 
the artist's palette. 

Massicot (the yellow oxide of lead). This yellow is 
obtained by roasting metallic lead, and can also be 
prepared by roasting white lead. 

It has been used from very early times, and is men- 
tioned among others by Leonardo da Vinci for use in 
flesh painting. 

The flesh of the "La Belle Ferroniere," by Leonardo 
da Vinci, in the Louvre, when examined under the 
microscope, apparently contains this pigment. 

Naples Yellow. A favourite yellow with the early 
oil painters is Naples yellow, made by heating together 
the leads of oxide and antimony. 

The original Naples yellow is supposed to have been 
a native pigment found on Vesuvius. Naples yellow, 
like massicot, has been removed from the artist's palette 
owing to the fear of darkening by the action of sul- 
phuretted hydrogen, and is now replaced by mixtures 
of yellow of the same tint. The older painters seem also 
to have used a yellow made by heating together the 
oxides of lead and tin. 

Indian Yellowy. Prepared from purree, which used 
to be used in water colour, is no longer obtainable, and 
there are no other permanent yellows available for the 




Greens, Blues, Lakes, Madder, and a List 
OF THE Permanent Pigments 


Chrome Green. There is apt to be some confusion 
o"v\'ing to the names given to a group of greens, some of 
which are permanent and some of which are fugitive. 
Chrome green is a mixture of Prussian blue and chrome 
yellow, and is not reliable. Oxide of chromium green, 
on the other hand, is one of the most permanent pig- 
ments on the artist's palette, and viridian, or, as the 
French call it, verte emeraude, the hydrated oxide of 
chromium, is also a permanent pigment. Viridian is 
prepared by heating together sodium bichromate and 
boracic acid and washing the product. Owing to its 
transparency, it will tend, if painted on too thick, in 
oil, to turn very dark in time. 

Cobalt Green. The cobalt greens are absolutely 
permanent, having been made at a white heat like 
cobalt blue, and it may be generally assumed that 
pigments made at high temperatures are permanent. 
Several tints of cobalt green are available. They are 
compounds of the oxides of zinc and cobalt, and have 
the same opacity as chrome yellow, and therefore will 
not alter nearly so much by the changes in the linseed 
oil as viridian. 



Emerald Green. Emerald green, which is a com- 
pound contained of arsenic and copper and which is 
very poisonous, is very Httle used by the artist to-day, 
but it is a remarkable pigment with a quality of tint 
contained by no other green. It is permanent, but if 
mixed either with cadmium yellow or vermilion, acts 
chemically upon these two pigments with the production 
of a deep brown colour. 

Malachite. Malachite green is a native carbonate 
of copper, and has been used through the whole history 
of painting and is a beautiful and permanent pigment, 
but wdth little staining power, and is little used to-day. 
Green bice, the artificial copper carbonate, which is 
of more doubtful permanence, is also very little used. It 
should not be mixed with cadmium yellow. 

Sap Green. Sap green is a notoriously fugitive pig- 
ment, but has been replaced by many artists' colourmen 
by a similar green prepared from a coal tar dye, which 
is much more permanent than the old sap green. It is 
a preparation from buckthorn berries by boiling down the 
juice with alum. 

Verdigris, the basic copper acetate, is one of the 
oldest pigments, its preparation by the corrosion of 
copper plates by acetic acid vapour being described by 
Pliny. It is rightly excluded from the artist's palette. 
I have described elsewhere how a permanent green was 
prepared from it in the Middle Ages by dissolving it in 
pine balsam. 


Prussian Blue. Prussian blue is obtained by the 
oxidation of a precipitate formed when sulphate of iron 
is mixed with ferro-cyanide of potassium. It is a very 
beautiful blue, and may be regarded as just on the 
margin between reasonably permanent and fugitive 



pigments. It may be included in the oil palette, but 
should not be used for water-colour painting, and is 
useless in Buon-Fresco, as it is destroyed by the lime. 
I was recently confirmed in my belief in its reliability 
in oil by seeing a ceiling painted with Prussian blue more 
than one hundred years ago, in which the colour was 
quite bright and good. Being a very transparent pig- 
ment at the blue end of the spectrum, it is seriously 
affected by the yellowing of linseed oil, and should 
therefore only be used for thin glazings and scumblings 
or mixed wdth a large quantity of white. 

Cobalt Blue. Cobalt blue, obtained by heating 
together in a furnace compounds of cobalt with alumina 
to a high temperature for a long time, is an absolutely 
permanent pigment, and can be used freely in water- 
colour and in fresco painting ; but being transparent 
and at the blue end of the spectrum, the same pre- 
cautions must be taken in its use in oil as in the case 
of Prussian blue. 

Ultramarine. The real ultramarine used by the 
old painters was extracted from lapis lazuli by mixing 
the finely ground mineral, after heating, with a mixture 
of resin and oil, leaving this for some weeks and then 
kneading the mass under water rendered slightly alkaline 
by wood ashes. In this way the blue is extracted, and 
separated from the other minerals present. This is 
a remarkably permanent blue of a very beautiful quality, 
and was much prized by the old painters. There seem 
to be certain conditions under which it turns grey, 
known as ultramarine sickness, but this is very rare, 
and probably is due to some accidental cause which has 
not yet been determined. In most cases the blue seems 
to last without any change. To-day we have artificial 
ultramarine made by heating together silica, soda, 
sulphur and coal, which while not so permanent as the 



natural ultramarine, seems quite reasonably permanent 
for ordinary use, having now been used by artists since 
its discovery early in the nineteenth century. It is not 
quite so transparent as cobalt blue, but is quite suffi- 
ciently transparent to get very dark in tone if painted 
on thickly in oil, all the more as a large quantity of oil 
is required to grind it. The same precautions should 
therefore be taken for using it in oil as in the case of 
cobalt blue. 

Cerulean Blue. This blue, made by heating com- 
pounds of tin magnesium and cobalt for a long time at 
a high temperature, is quite permanent, and has an 
opacity about the same as cobalt green and chrome 
yellow. It will therefore keep up its colour key in oil 
much better than any other blue. 

AzuRiTE. Azurite, or Azzurro della Magna, the 
beautiful native carbonate of copper which was used 
very largely between the middle of the fifteenth to 
the middle of the seventeenth century, is no longer 
obtainable, and the artificial variety of this carbonate 
of copper, blue bice, is a weak pigment which is little 
used to-day. At the same time its permanency and 
brilliancy, if properly used, can be seen in the portrait 
of Madame Pompadour by Boucher, in the National 
Gallery, Edinburgh. 

Egyptian Blue. This blue, manufactured and used 
in Egypt for over two thousand years, disappeared from 
the artist's palette somewhere between the second and 
the seventh century. If we had not cobalt blue, it 
would be worth while to revive its manufacture for use 
in Buon-Fresco. It is made by heating sand, copper 
carbonate, soda, and lime to a temperature between 
850° and 900° C. for several days. If the temperature 
is raised too high the green is obtained, which was used 
so much in Egyptian painting. 



Smalt. This beautiful blue, which was discovered 
towards the close of the sixteenth century, is a finely 
powdered cobalt glass. It was used very largely by 
seventeenth-century painters, and is to be found in the 
skies of Teniers. It is now very seldom used. It is 
quite a reliable pigment in oil. 

Indigo. Indigo, obtained from indigofera tinctoria, 
forms a very permanent dye, but as a pigment is quite 
unreliable, and should not be used. 

Violets. Cobalt violet and manganese violet — these 
two pigments, prepared from cobalt and manganese 
respectively by the formation of phosphates or arsenates, 
are quite reUable and permanent. 

The Lakes. We have now described all the pigments 
which are worthy to be included in the artist's palette, 
vnth the exception of the lakes. These pigments are 
obtained by dyeing the translucent precipitate of 
alumina obtained by adding soda to alum, with an animal 
or vegetable dye, and their preparation and method of 
manufacture is very old — apparently having been under- 
stood even in the time of Pliny. Their permanency 
depends entirely upon the dye selected, and conse- 
quently Dutch pink, in which the dye is Persian berries ; 
crimson lake and carmine, in which the dye is cochineal ; 
lac lake, in which the dye is obtained from the colouring 
matter in shellac and lakes from dye woods, are all 
fugitive and should not be used. Of all the dye stuffs, 
the madder root alone jields a dye sufficiently permanent 
for artists' purposes. It is difficult to know when madder 
lakes were first prepared, but they are at any rate as 
old as the fifteenth century, and where we find a brilliant 
lake which has resisted the changes of time in an old 
picture, we may take it to be madder lake. The madder 
lakes fade slowly when exposed to light, so that they 
cannot be regarded as of as high an order of permanency 



as a pigment like cobalt blue, but are sufficiently 
permanent for all practical purposes, both in oil and 
water colour. Being very transparent, they are slightly 
affected by the yellowing of linseed oil ; but not seriously 
so, owing to their being at the red end of the spectrum. 
The dyeing principle of the madder root, alizarin, has 
not only been extracted from the root, but is now made 
artificially. The lakes, prepared from alizarin are 
deeper and stronger than those obtained from the 
madder root, and seem to be equally permanent. To-day 
the word madder lake is very commonly used for lakes 
that may be made from the root or from synthetic 
alizarin or from blends of the two, but for practical 
purposes this need not trouble the artist. A more 
dangerous matter is that the word alizarin is now not 
only applied to crimson and red lakes, but also to green 
and blue lakes made from allied dye stuffs which, while 
all of a fair order of permanence, require careful testing 
before they can be assumed to be as reliable as the 
madder and alizarin crimson. 

I have identified as madder a beautiful lake on a 
manuscript in the Advocates' Library, Edinburgh 
(Speculum, Vitae, Christi, 18.1.7), which from internal 
evidence is known to have been inscribed between 1465 
and 1489. But the madder lakes are much older than 

Professor Russell, having found a pigment consisting 
of sulphate of lime dyed with madder among Egyptian 
pigments, succeeded in imitating it by boiling the madder 
root with lime and sulphate of lime. 

The question naturally arises why recipes for madder 
lakes are absent from mediaeval manuscripts. The follow- 
ing is, I think, the explanation : 

Madder was used for dyeing from early times, and we 
know for instance that from the thirteenth century 

G 97 


the dyeing with madder was an important industry in 
Marseilles. Now frequent references are made in the 
old recipes to the extraction of colour from dyed wool 
trimmings for making lakes. If, on the other hand, we 
attempt to make a bright lake from the madder root 
direct we get a dull lake, owing to the other dyes present. 
How this difficulty is avoided by makers of madder 
lakes I cannot say, as these are jealously guarded secrets 
and published recipes are of no value. 

I have myself succeeded in making fine lakes by first 
charring the madder root with sulphuric acid, so as to 
destroy everything but the stable alizarin group, and 
then dissolving out the dye stuff A\dth sulphate of 
alumina. Such methods were not available in early 
times, and therefore probably madder lakes were always 
prepared from the dyed wool trimmings. 

There are other pigments to be found in artists' 
colourmen's lists, some of which are quite useful and 
reliable ; but their name and number is legion, and it 
is sufficient for us to deal with those most commonly 
used and which the tests made by chemists and the 
experience of artists' colourmen have shown to be 

Permanent Pigments 

In the past. Captain Abney, Professor Russell, and 
Sir Arthur Church carried out experiments on the 
permanency of artists' pigments, experiments which 
are dealt with very fully in Sir Arthur Church's Chemistry 
of Paints and Painting. Since then Professor Eibner 
and others have carried out similar researches. 

In addition to these the English artists' colourmen 
have for many years not only obtained the valuable 
experience of manufacture and sale, but have carried 
out experiments in their own laboratories. The result 



of this cumulative experience and experiment is to be 
found in their catalogues. One firm limits the palette 
for sale to what they regard as sufficiently permanent 
pigments ; others di\dde the pigments into groups, 
according to their order of permanency. 

As a result of comparing these various sources of 
information I regard the following pigments as reUable 
for oil painting, namely : — 

White lead 

Permanent whites 

Zinc white 

Lamp or vegetable black 

Ivory black 

Charcoal black 

The Earth Coloues 

Yellow ochre 

Red ochre 

Raw sienna 

Burnt sienna 

Mars yellows 

Mars reds 

Venetian red 

Indian red 

Terre verte 

The umbers 


Cadmium scarlet 
Cadmium mid 
Cadmium pale (furnace 

Lemon yellow (barium 


Oxide of chromium 
Cobalt green 
Cobalt blue 
Cerulean blue 
Prussian blue (Antwerp 

Cobalt \iolet 
Manganese \-iolet. 
Madder and alizarin 

For water colour and tempera, white lead and Prussian 
blue must be excluded, and ultramarine if the egg 
medium contains acetic acid. For fresco only the earth 
colours and viridian, oxide of chromium, cobalt green, 
cerulean blue, cobalt blue, and cobalt and manganese 
violets should be used. 

In my opinion the most doubtful pigment in the whole 

list is cadmium yellow. 



It is evident that all the firms regard a mid cadmium 
as permanent, and that is the only one I have included. 
They are also satisfied as to the deep cadmiums, but are 
more doubtful as to the pale cadmium. 

As far as my o^vti experience goes, pale cadmiums are 
fugitive or permanent according to the method of manu- 
facture, while deep orange cadmiums have a tendency 
to become yellower. I have dealt with this more fully 
under cadmium yellow. 

Ultramarines require special consideration. They are 
largely manufactured for paper-makers, for whom they 
require to be of great stability to resist the action of 
alum, and a great deal of research has therefore had to 
be made by ultramarine manufactm'ers as to the con- 
ditions necessary for stability. It will be noted that 
they are accepted as permanent by all the artists' 
colourmen, yet it is known that the real ultramarine 
in the oil pictures by old masters sometimes suffer from 
" ultramarine sickness," the pigment becoming grey. 

This — it has been suggested by Van der Sleen — is due 
to the deposition of moisture containing sulphur dioxide 
in solution which in contact with the oil film becomes 
oxidized to sulphuric acid, and he has carried out many 
experiments to demonstrate this. 

It is evident that if ultramarine is to remain in the 
artist's palette the conditions for obtaining the most 
stable variety must be laid down. 

Prussian blue has always been regarded on the margin 
as a permanent pigment, but I know one example of a 
picture painted more than forty years ago with Prussian 
blue which is in perfect condition, and the Office of 
Works, in their repairs to the roof of the Great Hall at 
Hampton Court, proved that the wooden panels had 
originally been painted blue with azurite, had been re- 
painted with Prussian blue in the eighteenth century, 



and that this again had been covered with an imitation 
oak paint in the nineteenth century. The Prussian blue, 
which had been exposed for probably about a hundred 
years, was in perfect condition, which seems to justify 
its inclusion in an oil palette. 

Chrome yellow from lead may be regarded as being 
on probation. 

As stated in the beginning of this chapter, certain 
firms have added new pigments to their lists which they 
are satisfied are permanent, largely to replace fugitive 

These pigments are worthy of investigation, but in 
the meantime I have confined myself to pigments which 
have long been on the artist's palette. 

Professor Eibner has proved that ultramarine, ver- 
miUon, and cadmium yellow ground in oil can safely be 
mixed with white lead free from acetate. He condemns 
lead chromes, and has come to the conclusion that zinc 
oxide mixed with many pigments in water colour causes 
them to fade rapidly when exposed to light in presence 
of an excess of moisture. 




Before discussing further the properties of pigments 
and oils it is necessary to deal with certain fundamental 
optical principles. We are accustomed to regard a sheet 
of -window-glass as a transparent substance through 
which light passes without any obstruction. But a little 
consideration will enable us to reahse that obstruction 
does take place. 






Fig. 2. 

The window-panes of a railway carriage when the 
train is travelling at night act as very perfect mirrors 
reflecting the interior of the carriage, and a distant 
window-pane catching the rays of the sun flashes with 
a brilliance that can be seen for miles. 



The rays of light on passing from air to glass meet 
with obstruction and, in consequence, some are reflected 
while only a part is transmitted. If the incident ray hits 
the glass at an angle, the transmitted ray is bent out of 
its path on passing into the glass so as to shorten its path. 

The bending of this ray is called refraction, and the 
amount of bending is the measure of the obstruction 
suffered by the ray. 

If we proceed to experiment on various transparent 
substances such as water, oil, glass, and diamond, we 
shall find that the obstruction they offer to the passage 
of a ray of light through them varies, so that the pro- 
portion between the light which is reflected and the light 
which is transmitted also varies. 

In order to make this comparison we must be careful 
to arrange that the slope of the ray to the surface of the 
transparent substance, or angle of incidence, is the same 
in each case. The angle of incidence is defined as the 
angle between the incident ray and a perpendicular to 
the surface, and the angle of refractioi;! is defined as the 
angle between the refracted ray and perpendicular to the 
surface, and it has been found that, whatever the angle 
of incidence be, when light is passing from a vacuum into 
a given substance, such as glass or water, there is a definite 
mathematical relationship between the two angles which 
is defined as follows and is called the index of refraction. 

In mathematical language, if A is the angle of 
incidence, and B the angle of refraction, then the measure 
of the amount of refraction, which is called the refractive 
index, is obtained from the formula, the refractive index 

_ sin A 
sin B 

We can show experimentally that refraction takes 
place in the follomng way : Allow a ray of light to fall 



on a surface of water contained in a trough with parallel 
glass sides. We can then trace the paths of the rays 
o^^'ing to floating particles of dust in the air and in the 

We are all familiar %\dth the concept that light con- 
sists of tiny waves, and we have all seen waves of water 


^ RAY 






Through transparent liquid 
Fig. 3. 

striking a breakwater and being reflected from the 
surface, the reflecting waves running through the others 
out to sea. We may take as a rough analogy to a sub- 
stance like glass a pier made of piles \\'ith spaces between 
them. Some of the water waves will pass through, while 
others "vvill be reflected and thrown back. 

All that we need to understand for our present purpose 
is that a substance which has a higher " index of refrac- 
tion " is a substance which is offering more obstruction 
to the passage of the ray of light, and therefore a larger 
proportion of the ray will be reflected. 

We have now to ask ourselves what will happen when 
the ray of light is passing from one transparent substance 



of one index of refraction to another substance of another 
index of refraction. 

If, then, a ray of hght passes from a transparent sub- 
stance hke air with a low index of refraction to a trans- 
parent substance of a high index of refraction — that is, 
offering more obstruction to the passage of the hght, part 
of the hght will be reflected at the intervening surface. 
We can easily illustrate this with a glass trough used in 
the former experiment. The trough is now half filled 






rufractive: imdeix 
Fig. 4. 

with a liquid of a high index of refraction, and floating 
upon it is a liquid of a low index of refraction. We shall 
now see a reflection of a part of the ray where the two 
liquids meet. 

As a rule solid bodies have a higher index of refrac- 
tion than liquid, and consequently, if we again fill the 
trough with water alone, and suspend below the surface 



a horizontal sheet of glass, we shall again obtain a 
reflected ray where the glass and water meet. 

Let us now suppose that instead of a sheet of glass 
we introduce a transparent sheet of material with a low 
index of refraction, like gelatine ; we shall then get a 
comparatively feeble reflection where the water and 
gelatine meet. 







Fig. 5. 

Or if instead of remo%ing the glass we replace the water 
by a series of liquids of higher and higher refractive index, 
the reflection will be feebler and feebler until, if we finally 
use a liquid of the same refractive index as glass, there 
will be no reflected ray at the meeting of the Uquid and 
glass at all. 

These experiments may seem remote from the subject 
of oil paintings, but they have very direct practical 
application, as we shall presently see. 



To sum up our conclusions, when a ray of light is 
passing from a transparent substance of low refractive 
index to a transparent substance of high refractive index, 
light is reflected at the boundary, and the amount of 
light reflected will be the greater the greater the differ- 
ence between their respective refractive indices. 

We shall now return to the glass sheet in order to carry 
out an experiment of the greatest practical importance, 
but before carrying out this experiment let us hold the 
glass plate so as to look at the edge of the plate. The 
hght is now travelling to the eye through a considerable 
thickness of glass and the glass appears green in colour. 
Therefore, besides reflecting and refracting the light, the 
glass is absorbing a certain kind of light and so appears 
green. We shall have more to say about this presently. 

Our next experiment is to break the glass sheet to 
pieces and grind it into a fine powder. This powder is 
white and opaque. If we had not seen the glass as a 
continuous sheet, we would never have thought of it as 
a transparent substance, and yet each separate particle 
of the powdered glass, if examined under a lens, is seen 
to be transparent. The explanation is that the light 
reflected from a million facets is scattered, and it is 
scattered light which produces the effect of opacity. It 
is by scattered light that we are able to see objects, and 
it is with scattered light that the artist has to do. 

We shall now pour the glass powder into a glass vessel 
and then fill up this vessel with water. Water has not 
so high a refractive index as glass, but its index is much 
higher than that of air. As the liquid rises, the glass 
powder no longer reflects so much light. It becomes 
partly translucent, and this although the liquid has not 
so high a refractive index as glass. If the liquid had the 
same refractive index, the glass would be almost in- 
visible. We already know the explanation. Reflection 



takes place when light is passing from a transparent 
substance of lower to one of higher refractive index. 
Though water is not of so high refractive index as glass, 
it is very much higher than air, and as we see the glass 
by scattered reflection, if the reflection no longer takes 
place, the glass disappears, just as in the story by Wells 
of the invisible man ; he became invisible by reducing 
the refractive index of all his tissues to the refractive 
index of air. 

We can at once proceed to make some very important 
practical applications of these experiments. We are 
accustomed to regard some pigments as transparent or, 
at any rate, translucent, and others as opaque. 

It is a question which we shall have to discuss later 
whether there is any such thing as an opaque pigment. 
We shall, at any rate, find that many pigments usually 
regarded as opaque are transparent in a highly refractive 
medium. In the meantime we shall confine our re- 
searches to the bearing which our observations have on 
white lead and zinc white, so as to avoid the complica- 
tions of colour which we shall have to consider presently. 

If white lead is ground in a liquid of highly refractive 
index like bromonaphthalene, it becomes a translucent 
pigment of a low-toned greyish white ; and if the ground 
pigment is flooded ^vith bromonaphthalene and exa- 
mined under the microscope by transmitted light, it is 
seen to consist of transparent crystals with a high 
refractive index, gUttering hke diamonds. White lead, 
therefore, is an extreme case of our powdered glass, and 
the amount of white light it will reflect will depend on 
the refractive index of the liquid in which it is immersed. 

It reflects most light as a dry powder in air and reflects 
less light when ground in a medium like egg or size, and 
still less light when ground in oil, which has a higher 
refractive index than either egg or size. 



If zinc white is examined in the same way, it is also 
found to consist of transparent particles, and therefore 
behaves in the same way as white lead in media of 
varying refractive index. 

We now come to a very important practical question. 
I am proposing to put a thin coat of oil over a white 
ground from which I wish to secure the maximum bril- 
liancy by the reflection of scattered light. I have 
therefore painted the ground with white lead or zinc 
white mixed %nth a low refractive medium like size. 
What would be the effect of varnishing the surface with 
oil ? The reply is, that if the surface is absorbent, that 
is, if the pigment is not effectively coated with size, the 
oil ^^^ll soak in, ^^'ith degradation of tone. But if the 
surface is non-absorbent, there will be so little degrada- 
tion of tone as to be hardly visible. 

Therefore to obtain the highest illumination through 
a layer of oil we must mix the white lead or zinc white 
with a medium of low refractive index and introduce 
sufficient of the medium to produce a non-absorbent 

We shall now proceed to make another practical 
application of our experiment. 

Every artist knows that if an under-painting of 
definite shape is painted out it slowly becomes visible, 
appearing like a ghost through what is painted above. 
Such pentimenti can be detected in many pictures. It 
is evident that, while this coming through of the under- 
painting is only visible where the under-painting had a 
definite shape, the same increased translucency of the 
surface layer must be taking place all over the picture. 

If we paint out a board mth black and white squares 
and then paint on layer upon layer of zinc white or white 
lead in oil, until the black and white squares are no 
longer \'isible, and then expose the whole to light, in a 



few months the black and white squares begin to show 
out again. 

It is a commonplace of artists' experience to find that 
Venetian and Indian red " come through " a surface 
painting. These pigments, as we shall find, are among 
the most opaque we possess. 

Let us see if we can find an explanation of these 
phenomena. We know that zinc white and white lead 
are transparent pigments, and that they will become 
more and more transparent if immersed in a medium of 
higher and higher refractive index. If, then, the linseed 
oil film increases in refractive index, as it gets older we 
should expect the results that are found in practice. 
Measurements of the refractive index of a linseed-oil 
film shows that it rises in refractive index while and after 
drying, and this change probably continues through 
many years. While the change in translucency may in 
part be due to chemical change, the most important 
cause is probably the change in refractive index of the 
dried oil film. 

We should therefore expect to find, as we do find, 
that pigments like white lead and zinc oxide will become 
more transparent with age, and therefore, if we have 
darker paint below them, the picture wUl go dovm in 
tone ; but if we have lighter paint below them, then, 
if the oil remain colourless, the picture would tend to 
go up in tone. We shall now proceed in the next chapter 
to investigate the question of colour. 




In the last chapter we described some experiments with 
a sheet of glass. For the experiments we are now about 
to make we shall replace the sheet of glass by a prism 
of glass and allow a ray of white light to fall upon it as 
shown in the diagram. 

Fig. 6. 

This ray will be refracted both on entering and leaving 
the prism ; and, having arranged matters so as to have 
sloping sides, the ray will be bent in the same direction 
both on entering and leaving the prism, and we shall 
find that, in addition to refraction, something else is 
taking place. 

Besides being refracted, the ray has been opened out 
fanlike into a band of colour beginning with red, passing 
from red to orange, orange to yellow, yellow to green, 
green to blue, and blue to violet. Dispersion as well as 
refraction has taken place. This band of colour is called 
the spectrum. 



Two explanations might be given of this result — one 
that the glass had somehow changed the nature of white 
light ; the other that white light itself is a complex of 
different coloured lights which we have succeeded in 
separating from each other. 

In order to decide which of these two things has 
happened we shall make two experiments. We can 

Fig. 7. 

catch our coloured band in a second prism placed the 
opposite way, in which case our coloured band is closed 
up again and appears as a ray of white light. 



Tig. 8 

Or we can catch our coloured band on a row of little 
mirrors so placed as to close it up again, when a ray of 





white light is again obtained. We are therefore justified 
in concluding that white light is in reality composed of 
these different coloured lights. 

We have already spoken of Hght as composed of tiny 
waves which are obstructed in their passage through 
glass. Experiment has demonstrated that a ray of 
white light is a complex of waves of different wave- 
lengths and that each wave is refracted at a different 
angle according to its wave-length, the long red wave 
being refracted least and the short violet wave being 
refracted most, so that on passing through the prism 
they are bent apart and separated from each other. 
These waves of different wave-lengths produce different 
colour sensations, which we name red, orange, yellow, 
green, blue, and violet. 

We are now on the way to a simple explanation of 
why different substances have different colours. Sup- 
pose instead of using a piece of colourless glass, as we did 
in the former experiment, we use a piece of red glass, we 
shall find that, in addition to reflection and transmission, 
the red glass absorbs, destroys, or obliterates light of 
certain wave-lengths — namely, the blue, green, and 
yellow, and only allows the red waves to be transmitted. 

All colour is contained in white light. Some substances 
reflect all the light, and we call them white. Some 
absorb the light, and we call them black ; and some 
absorb certain coloured waves and transmit the others, 
and we call them red, yellow, green, blue, and so on. 

Let us next try to understand what happens when 
two or more pigments are intimately mixed together. 
In the first place, we shall select pigments which are 
transparent or, at any rate, translucent in oil, such as 
ahzarin red, cobalt yellow, viridian, and cobalt blue. 
If we mix these together in the proper proportions we 
sliall obtain a black. 

H 113 


The explanation is simple : the white light is being 
filtered and refiltered, sinking deep into the transparent 
mass of oil and pigment, transmitted and reflected and 
retransmitted, and as it passes through each pigment it 
is robbed of some of its coloured rays — ^the cobalt blue 
robs it of all but blue and violet, the alizarin of all but 
red, the cobalt yellow of all but yellow, the viridian of 
all but green, till nothing is left. 

In our next experiment we shall select two pigments — 
a pale chrome, or pale cadmium yellow, and Prussian 
blue. On mixing these two pigments we obtain a green. 

In order to explain this result we must find out what 
part of the spectrum each of these pigments reflects. 

This can be done by throwing the spectrum band of 
colour on the pigment thickly painted out and observing 
what part of the spectrum it reflects. As a result of this 
experiment we find that what we call a pure pale yellow 
pigment also reflects a great deal of white light, and 
more especially the whole of the spectrum from red to, 
and including, green, and that a blue, like Prussian blue, 
also reflects a good deal of the green. 

When mixed together there will be a certain amount 
of light reflected from the surface particles of both pig- 
ments, which we shall find later on gives us a cold grey ; 
but the greater part of the light, penetrating into the 
mass of mixed pigment and being reflected and trans- 
mitted from particle to particle, loses all the blue-violet 
by absorption by the yellow, and all the red-yellow by 
absorption by the blue, and what is left is the part of 
the spectrum common to both pigments — namely, green. 

If we take an orange-tinted yellow like mid-cadmium, 
and a purple-blue like cobalt blue, we get our mixing 
a dirty brownish green, because we no longer have pig- 
ments which are reflecting much of the green band of 
the spectrum. In fact, the colour obtained from a 



mixture of pigments is always very largely a residue 
effect, each pigment absorbing its complementary colour 
in the spectrum. We shall have to go much more fully 
into this matter of colour and colour sensation later on. 
At this stage I propose to deal only with certain aspects 
of it. 




re:d yellow grleln blue: 





Fig. 9. 

This diagram shows quite clearly how it is that a 
mixture of blue and yellow produces green. The 
shaded portions indicate the colours of the spectrum 
which are destroyed bj- each pigment. Thus we see 
that the green light alone escapes. 



We now come to a fundamental experiment of the 
greatest importance if we are to understand all that 

In the last chapter I described an experiment with 
powdered glass. We found that colourless, transparent 
glass, when powdered, gave us an opaque white powder, 
but on immersing it in a medium of a higher refractive 
index than air it became translucent. This we found 
was due to the fact that when light passes from a medium 
of one refractive index to another of a higher refractive 
index a great deal of light is reflected at the boundary 
between the two media, and as they approach nearer 
and nearer in refractive index, less and less is reflected 
until when they have the same refractive index no light 
is reflected at all, but it is all transmitted. 

We are now dealing not only with reflection and 
refraction, but also with absorption of certain coloured 
rays and reflection and transmission of the comple- 
mentary colours. Let us try to understand what happens 
when a ray of white light strikes a piece of ruby glass 
at an angle. In the first place, the amount of absorption 
depends on the depth of penetration. 

A very, very thin sheet of ruby glass would hardly 
appear red at all, for as the light passes through 
layer after layer more of the blue-green end of the 
spectrum is absorbed, and the colour gets deeper and 

It follows from this that the light that is reflected 
from the surface where air and glass meet vn\\ consist 
very largely, if not entirely, of white light. The light 
will not have penetrated sufficiently for much absorption 
to take place. There may have been some absorption, 
so that the reflected light may be tinged with red, but 
roughly, for our present purpose, we can regard it as 
consisting largely, if not entirely, of white light. In the 



case of scattered light from a broken surface the re- 
flected light ^vill contain a larger portion of the pre- 
dominating colour mixed with white. For in the case 
of the rough surface necessary to scatter the light, some 
of it is transmitted before it is reflected. 

This being understood, let us proceed to repeat our 
former experiment with colourless glass and grind our 
deep ruby glass into a powder. The result is that we 
obtain a pale pink opaque powder, which, the finer it is 
ground, approaches nearer and nearer to white. We 
have multiplied a miUionfold the surface reflections, and 
these, as we have seen, consist largely of white light, 
and so we obtain a pale pink opaque powder. Now 
mix this powder with a liquid of moderately high re- 
fractive index — linseed oil will do for the purpose — and 
at once we obtain a deep red translucent pigment rather 
like madder or crimson lake. 

In order to understand this, we must go a little further 
back to our experiment with the glass plate immersed 
in a trough of water, but instead of one glass plate we 
shall now place three plates of ruby glass one below the 
other — A, B, and C. In the first place, let us suppose 
the plates immersed in a liquid of very low refractive 
index, so that there is a great deal of white light (2) re- 
flected from the surface of the first plate. The red ray (3) 
transmitted through the first plate will be partially 
reflected by B, giving a reflected red ray (4) which will 
mingle with the white light (2). The red ray which 
passes through B will again be partially reflected by C, 
and so on. 

If most of the light has been reflected by A as white 
light, the red rays from B and C \dll be diluted with a 
great deal of white light. 

If we now replace the liquid of low refractive index 
by a liquid of high refractive index, very little white 



light will be reflected by A and most of the light will be 
transmitted through A, with reflections of red light by 
B and C, and so on, these red rays being very little 
diluted with white and therefore giving it a deep- 
coloured reflection. 








Fig. 10. 

We have repeated our old experiment with a variation 
by introducing absorption of certain portions of the 
spectrum, as well as reflection and refraction. In this 
description the internal reflections at the glass air 
surfaces are omitted, as they are common to both 




We have next to ask ourselves, Are the pigments 
which we use transparent or opaque ? We have already 
found white lead and zinc oxide to be transparent, but 
how about the others ? Many we know to be trans- 
parent or, at any rate, translucent — namely, the lakes, 
cobalt yellow, viridian, Prussian and cobalt blue. In 
order to test whether any more of them are transparent, 
the obvious plan is to immerse them in liquids of higher 
and higher refractive index, so as to get rid of surface 

White lead and zinc oxide become dirty grey, trans- 
lucent pigments in bromonaphthalene, and are seen 
when immersed in this liquid and examined by trans- 
mitted light under the microscope to be composed of 
transparent crystals. 

Let us, then, mix some of our best known bright pig- 
ments in bromonaphthalene and observe the change of 
tints produced by cutting out some of the surface 
reflections. For this purpose the pigment is best piled 
up in a little heap and then wetted with bromonaphtha- 
lene so as to get the full value of the internal reflections 
and transmissions. 

Vermilion and the madders do not change very much 
in tint. Bright French yellow ochres become a brownish 
yellow ; raw sienna a deep bro-\Mi ; pale chrome and 
cadmium yellows brownish orange ; the green pigments 
much darker and deeper ; and the blues almost 

We learn some very interesting facts from these ex- 
periments. The mingling of white light with the red 
pigments makes them brighter, but does not alter their 
tint very much, and the same is true of greens and 

In the case of yellows the eye is capable of obtaining 
the sensation of yellow from a small portion of the 



spectrum — namely, the D double line of sodium ; but 
the yellow sensation is also e^^dently produced by a 
mixture of orange and white light which the pale yellow 
pigments reflect. 

The next most obA-ious experiment is to examine the 
pigments ground in bromonaphthalene under the micro- 
scope. Vermilion, raw sienna, chrome yellow, cobalt 
green, ultramarine, and cerulean blue all prove to be 
transparent ; French light yellow ochre, cadmium 
yellow, and Venetian and Indian reds being opaque. 

It is evident from their change of tints and deepening 
of tone that they also are probably transparent in a 
liquid of sufficiently high refractive index. 

If we now grind these remaining pigments in methy- 
lene iodide, saturated %\-ith sulphur, which as a refractive 
index of about 1-8, yellow ochre and cadmium yellow 
are seen to be transparent, and we are only left ^\'ith 
Venetian and Indian red. 

Doubtless a liquid of sufficiently high refractive index 
would resolve these also, as we know the crystalhne 
haematite in sufficiently thin layers is transparent, and 
the blacks and bro%A'ns will also prove to be transparent, 
highly absorbent substances. 

The opacity of a pigment is due to the amount of 
absorption, of which black is a typical example, also 
to the refractive index, to the size of the particles 
and to the amount of internal reflection within the 

The result of this rough analysis of the principal 
bright pigments will enable us to arrange them in a 
table of a relative opacity, or with more exactness relative 
refractive index, the practical importance of which will 
appear later. 

We can now explain many phenomena -^ith which the 
painter is famihar. 



When pigments are mixed with water, or water to- 
gether with size, gum, or egg, they are comparatively 
deep in tint, but dry, very much paler. This is because 
the spaces occupied by the water of high refractive index 
are replaced by air of low refractive index, and therefore 
much more white light is reflected from the surface layer. 

Pictures painted in size or egg are much more brilliant 
than pictures in oil, because of the low refractive index 
of these media. 

A very important practical question now requires to 
be answered. We found in Chapter VHI that a layer 
of colourless oil did not effect the brilhancy of a pigment 
painted out in size or egg if the ground was non- 

Is there any way in which we can restore the brilhancy 
of a pigment ground in a liquid of high refractive index ? 
There is a way. If painted out very thin on a white 
solid, non-absorbent surface, the light passing through, 
reflected back, and passing through the thin layers of 
translucent or transparent pigment restores its full 
brilliancy. Cobalt blue, for instance, heaped up and 
wetted with bromonaphthalene appears almost black ; 
spread thinly on a white porcelain plate, it at once 
becomes again a brilliant blue. 

Finally, it is obvious that if, as we have found, linseed 
oil increases in refractive index with age, all pigments 
will deepen in tone, the most transparent being most 
affected, and the most opaque least affected. 

In another chapter we shall discuss the cause of the 
yellowing of the oil and how much can be done to prevent 
it, and how far different oils vary in the extent to which 
they turn yellow. Here we are interested only in the 
fact that oils do yellow and the optical results that will 
follow. After the experiments already made, and the 
facts ascertained, we can decide what would be the 



effect on different pigments. The simplest way of con- 
sidering this is to imagine a Httle of a transparent 
brownish yellow pigment mixed in throughout the 

In order to understand what will happen, let us take 
as an example a pigment like cobalt blue. Cobalt blue 
when examined under the microscope is seen to consist 
of transparent blue particles. The light will therefore 
penetrate to some depth before it is returned to the 
surface by internal reflection. If the blue particles are 
floating in a colourless medium, this will not matter ; 


FlQ. 11. 

but if we suppose them to be floating in a brownish 
yellow medium, which consequently absorbs the blue 
end of the spectrum, we may get complete absorption. 
The cobalt blue absorbs the red-yellow end, and the 
brownish yellow absorbs the greenish blue end, and so 
between them we get complete absorption, and the 
cobalt blue appears to be black. An opaque blue like 
cerulean blue does not suffer so much from the yellow- 
ing of the oil because there is more surface reflection. 
Pigments at the red end of the spectrum will not suffer 
so much because they are absorbing very much the same 
part of the spectrum as the brownish yellow oil. Trans- 
parent pigments at the green and blue end of the 



spectrum, if painted in mass, are bound to be degraded 
by the yellowing of the oil. If thinly glazed or scumbled 
on a brighter foundation, or mixed with white, which 
has the same effect, the particles of white reflecting the 
light back through the transparent particles of pigment, 
they will then keep nearly their full value in the picture. 





FlQ. 12. 

From the experiments made with different pigments 
in liquids of different refractive index we are now in a 
position to draw up a table which w411 give us a rough 
measure of the effect of the increasing refractive index 
and the yellowing of the oil on pigments. White and 
black and dark brown pigments like umber are omitted, 
though as an actual fact the blackness of black varies 
according to the nature of the black and the refractive 
index of the medium, all blacks reflecting some white 
light. The whites we have already studied. The nearer 
a pigment is to the blue end of the spectrum, and the 
more transparent it is, the more it will be affected. I 
have selected for my table the pigments most commonly 



£^used which are from the chemist's point of view fairly 






VeneMan & 
Indidfi Red 




Lighl" French 


Clirom callows 

Oxide of 
Chromiiuu & 

Cerulean Blue 

Burnt Sienna 

Raw Sienna 


Madder & 



Coball" Yellow 


Prussian Blue 
Cofcalh Blue 



OIL- ^ 

Fig. 13. 

As nearly as possible the position of the pigments has 
been placed so as to indicate the effect of the alteration 
in the oil in the pigment if painted thickly. For instance, 
cobalt yellow being a yellow is not so much affected, 
but being very transparent, suffers more than an opaque 
pigment. Cobalt green being nearer the blue end of the 



spectrum is affected, but being also very opaque does 
not show much change. Pigments Hke cadmium scarlet, 
Venetian and Indian red ^vi\\ remain practically un- 
changed, and cadmium yellow will be changed less than 
chromes. Raw sienna, owing to its transparency, will 
darken, but owing to its colour is not very seriously 
affected. There is no highly opaque blue, the nearest 
approach being cerulean blue. The madders and 
alizarin, cobalt yellow, viridian, Prussian blue, cobalt 
blue, and ultramarine should only be used for glazing 
and scumbling on a light solid background or for mixing 
with white. 

All this has nothing to do with the permanence of the 
pigment from a chemical point of view. Cobalt blue, 
for instance, is of the highest order of permanence, and 
is therefore of the utmost value in fresco, water colour, 
and tempera. We are dealing here with optical effects 
of the oil in altering the tint and brilliancy of the 

Finally, it should be written up above the door of 
every atelier : " Remember white lead and zinc white 
ground in oil are transparent pigments." 

Let us now return to our experiment for the making 
of black by a mixture of alizarin, cobalt yellow, viridian, 
and cobalt blue in oil. If instead of taking these pig- 
ments as ground in oil we had taken them ground in size, 
or, still more, if we had mixed the dry powder, we would 
have obtained a grey, fairly deep for size, and much paler 
in the case of the dry pigment. 

Or if instead of these pigments we had selected pig- 
ments which are opaque in oil in comparison, such as 
vermilion, cadmium yellow, cobalt green, and cerulean 
blue, we would have got a grey. It is not difficult after 
what we have learnt about the reflection and trans- 
mission of light by pigments to understand these effects. 



If the particles of the pigment are lying side by side, 
and if they are sufficiently opaque to reflect most of the 
light falling on them, then, though this light is deeply 
tinted with their specific colour so that we may assume 
for our present purpose to be red, yellow, and blue light, 
we already know that the combination of coloured light 
produces white light ; consequently the rays of light 
from such a mixture reflected from the surface will 
produce the sensation of white light on the eye, and 
consequently a mixture of pigments covering the range 
of the spectrum will appear light grey to black, accord- 

rcd yeillow crelckj bluc 

Fig. 14. 

ing to their transparency and the refractive index of the 
medium in which they are ground. 

We can now understand why we get quite different 
effects if we paint little dabs of pure pigment side by 
side instead of intimately mixing them. We get the full 
value of the light reflected by each pigment, the coloured 
rays being blended in the eye, and consequently a much 
more brilliant result is obtained than by intimately 



mixing, by which large portions of the complete spectrum 
are absorbed . 

One curious and interesting result is that this 
method of painting does not produce a sensation of 
green. Blue and yellow painted side by side produce a 
cool grey. In the case of other pigments, the tints 
obtained by the two methods, while different in bril- 
liancy, belong to the same order of colour — a brilliant 
purple replacing a deep dull purple, and so on. Only in 
the case of blue and yellow a green can only be obtained 
either by using a green paint or mixing a suitable blue 
and yellow together. 

The connection between this interesting fact and the 
Young Theory of Colour Sensation will be found fully 
discussed in books on Colour. Here we are strictly con- 
fining ourselves to experimental facts. 

From what we have already learnt, the meaning of 
a complementary colour is sufficiently ob\dous. If any 
portion of the spectrum is selected, the rest of the 
spectrum, which is required with the first portion 
selected to produce the sensation of white light, is the 
complementary colour. There are thus an infinite 
number of complementary colours. Two complementary 
colours laid side by side mutually reinforce each other. 
Sir Ai'thur Church gives the following list of useful com- 
plementary colours in his book on Colour : — 

Red — Blue-Green. 
Orange — Turquoise. 
Yellow — ^Lapis Lazuh. 
Yellow-Green — Violet. 
Green — Purple . 
Green-Blue — Carmine . 




Linseed Oil 

Linseed oil is prepared by grinding, heating, and press- 
ing the seeds of the common cultivated flax, Linuvi 
usitatissimum. As it comes from the press it requires to 
be refined. This can be done in various ways. The 
oldest and the most satisfactory manner for artists' 
purposes is to expose it to light and air in covered glass 
vessels. A variation of this method is to float it on salt 
water, introducing as well a certain amount of sand. 
Large glass flasks are filled one-third of salt water, one- 
third of oil, and are then loosely corked and placed 
outside. Every day for the first two or three weeks the 
contents are \'igorously shaken up. The oil is then left 
for a few weeks to clarify and bleach. By this process 
mucilaginous and albumenoid substances are removed 
from the oil, and the final product, pale and clear, dries 
quite quickly enough for artists' purposes. 

O-^ing to partial oxidation, this oil has a somewhat 
high acid value. 

The same result is obtained on a large scale by the 
addition of a small quantity of sulphuric acid, which 
chars and removes impurities, and subsequent washing. 
Other methods of refining and bleaching are also known. 

Linseed oil consists principally of the glycerides of 
three fatty acids — linolenic acid, hnoleic acid and oleic 



acid linked together. On exposure to air oxygen is 
absorbed and the oil oxidised, being converted into a 
transparent leathery and insoluble substance which is 
called linoxyn. 

During this process the oil film is not only absorbing 
oxygen and therefore increasing in weight, but is also 
losing certain volatile products of the oxidation, thus 
losing weight. If a thin film of the oil is painted out on 
glass and weighed from time to time, it will be found to 
increase in weight in passing from liquid to sticky, and 
then from sticky to surface dry. It now begins to lose 
in weight, the rate of loss slowly diminishing. 

The complete oxidation of a film of oil is a very slow 
process, a film four hundred years old giving reactions 
which show that it is not yet complete. It is doubtless 
for this reason that the picture -cleaner finds he can dis- 
solve off fresh painting from older painting underneath, 
the older oil resisting the action of the solvent. The 
dangerous time for cracking is just after the film has 
become surface dry, as it now has a skin on the surface 
and is at the same time losing weight and shrinking. 

The principal agent in promoting the oxidation of a 
film is the linolenic acid. While the chemical changes 
produced by the oxygen are going on, which result in 
the drying of the film, they are followed by changes 
produced by moisture which result in the yellowing of 
the film. This yellowing is unfortunately the property 
of linolenic acid itself, and is not due to any impurity 
in the oil. Yellowing is promoted by certain substances, 
such as lead driers, explaining, as I have formerly 
suggested, the excessive darkening of some of the paint 
films painted out by Mr. Holman Hunt. 

As yellowing is produced by moisture, protection of 
the oil paint from the introduction of moisture, both 
from the back and the front, is important. The linseed 

I 129 


oil film itself is easily permeable by moisture. The films 
left by the spirit varnishes, like mastic, are not. I have 
already advised protecting the back of the canvas with 
tinfoil or with beeswax and resin, which is imperme- 

The yellowing of the oil takes place most rapidly in 
the dark. On exposure to light in a window the oil is 
bleached. On replacing in the dark, it again yellows. 
In the diffused light of a room a certain amount of 
yellowing takes place. What ultimately happens, 
whether the yellowing reaches a maximum in course of 
time and then begins to bleach — in fact, the entire 
relationship between yellowing, light, and time — is well 
worthy of investigation. It has been assumed from 
certain traditions and the Rubens Letters that it was the 
custom of the early painters in oil to put their pictures 
out in the sun while drying. This was probably often 
done ; but we now know that exposure to sunlight while 
drying is very injurious to a linseed-oil film. 

If linseed oil is heated and exposed at the same time 
to oxidation by blowing air through it, an oil is obtained 
which dries more quickly than raw oil. This drying is 
still further promoted by dissolving in the oil certain 
substances which are called driers and hasten the drying 
process, such as sugar of lead, litharge, and compounds 
of cobalt and manganese. These drying oils are seldom 
required by artists. Sometimes a little is added to 
hasten the drying. A drying oil with a lead drier should 
in this case be avoided, manganese or cobalt driers being 

The preparation of thickened oils by boiling was 
known in mediaeval times. Eraclius describes the 
boiling of the oil with calcined bones, which would help 
to remove water from the oil ; and Cennino Cennini 
describes the boiling of an oil till reduced to one-half 



its volume for preparing a mordant for sticking on gold 
leaf. For painting purposes he advises bleaching the 
oil in the sun. This boiling of the oil will produce two 
changes — partial oxidation and at the same time what 
the chemist calls polymerization, or the grouping of 
the oil molecules into larger molecules. If linseed oil is 
boiled with the exclusion of air, polymerization alone 
takes place, and a thick sticky oil is obtained, known as 
Stand oil, which has long been prepared in Holland. 

It is slow in drying unless driers are introduced ; 
rather dark, sticky, and flowing, so that it has all the 
faults which a painting oil can have, but it has one 
advantage — it yellows much less than linseed oil pre- 
pared in any other way, and Professor Eibner suggests 
it may have been used by the early painters in oil. It 
also probably produces the most permanent film of any 
oil. It is a constituent of the enamel used by the house- 

Many years ago, as stated in Chapter V, I tested the 
assertion of the house-painter that black and white 
squares when painted out with coats of white lead until 
they became invisible would show again in time, and 
I have recently confirmed this experiment. Professor 
Forbes tells me that he has been for some years 
making various tests on similar lines with a view to 
showing the extent to which lower coats of paint 
gradually become more visible. There are three 
possible causes — slow interpenetration of the pigment 
by the oil, formation of soaps with increased translucency, 
or a change in the refractive index of the oil itself. 
If it is due to slow penetration of the oil, or formation 
of soaps, this should surely take place when the pig- 
ments are ground in oil in tubes. I have never seen any 
indication of this. Increase in the refractive index could 
only result in increased transparency if the pigments 



were themselves transparent. As I have shown in 
the chapter on Oil Painting, the greater number of 
the pigments used by the artist, including white lead, 
are transparent, or, at any rate, translucent, and it 
becomes necessary to investigate whether the linseed 
oil film increases in refractive index with time. In 
order to test this a film of linseed oil, cold-pressed and 
sun-refined, obtained from a well-known artists' colour- 
man, was spread on the glass of a Herbert Smith re- 
fractometer, and the refractive index measured with 
sodium light. 




I 2 3 4-5678 DAYS 


Fig. 15. 

It is evident from this experiment that the refractive 
index of a linseed oil film is slowly rising, as we know 
the chemical change continues in the films for many 
years, the probability is that the refractive index con- 
tinues to rise as well. Experiments continued for many 
months show the refractive index to be still rising. 
After nine months it has reached 1-500. Nut and poppy 
oil will doubtless behave in the same way. 



Walnut Oil 

The kernel of the wahiut contains a drying oil, the 
use of which has been known since the time of ^tius 
of the fifth century, who advises varnishing wax pictures 
with it. Linseed oil is the only oil mentioned by Theo- 
philus, but there can be little doubt that the use of 
walnut oil for painting is very old. Either oil, linseed or 
walnut, seems to have been regarded as satisfactory. 
Vasari mentions both, and only remarks that walnut oil 
does not yellow so much as linseed oil. It can be ob- 
tained from the walnut kernel by grinding and pressing, 
or it can be extracted by boiling with water, the oil 
rising to the top ; or by dissolving it in a suitable sol- 
vent. It has long been used as an edible oil ; but is not 
used to-day fot the grinding of artists' pigments as a 
general rule, and is difficult to obtain commercially. 
It is now obtainable from America, being prepared pure 
under conditions laid down by the American Walnut 
Growers' Association. It dries very slowly, but if 
exposed to light or air over water it can be obtained 
very pale in colour and drying quite as quickly as the 
linseed oil prepared for painters' use. 

According to Professor Eibner's experiments, while 
this oil does not darken nearly so much as linseed oil, 
it has two defects — its curve of drying is practically the 
same as the curve of drying for poppy oil shown on 
page 135. That is to say, it loses weight very rapidly 
after it is surface dry, and is therefore very likely to 
crack. It doea not form so insoluble and permanent a 
film as linseed oil. This is due to the fact that while 
very similar in composition it contains a good deal less 
linolenic acid. 



Poppy Oil 

Poppy oil, which is expressed from the seed of the 
opium poppy, known from classical times, was first 
brought into use early in the seventeenth century, and 
we have reason to believe was very largely used by the 
painters of the Dutch school, both to paint with and to 
grind with their pigments. It is used to-day by the 
artists' colourman for the grinding of flake white, 
because it is so pale in colour. Like walnut oil, it does 
not darken much in time, but also, like walnut oil, it 
is defective in the amount of linolenic acid it contains, 
and is much more likely to crack than linseed oil and 
does not form nearly so insoluble a film. Its composition 
is similar to linseed oil, but like nut oil contains less 
linolenic acid. 

Other drying oils are known, but the only other one 
of importance to artists is Tung oil, prepared from the 
seed Aleuntes cordata. This is a drying oil, which has 
always been used in China and is now exported in large 
quantities to Europe. It owes its drying qualities to 
quite a different set of fatty acids to those found in 
linseed oil. It is a reliable drying oil, but unfortunately 
yellows just like linseed oil. It has the property of 
drying flat, and makes an excellent emulsion with size. 

In conclusion, the question as to which of these oils 
it is safest for the painter to use is a very difficult one. 
Linseed oil has the defect of yellowing more than the 
other two, but it is less likely to crack in drying and 
produces a more durable film. 

The following diagram, which gives the general form 
of the drying curve for linseed and poppy and nut oil, 
brings out clearly the results of Professor Eibner's 
researches. In each case a very thin film of oil on glass 
was weighed at intervals. In the case of linseeed oil 


Thej^e are microphotographs, in plan and in section, of a cracli in an oil painting. The shrinliirg 
of the layerof paint above the priming, causing a crack with straight sides, is clearly shown. The 
pull on the priming by the shrinking layer of paint is also shown, the priming having been drawn 
out. Incidentally, it is very difficult for the forger of Old llastersto produce artificially cracks 
■which will stand examination underthe microscope and comparison with genuine cracks produced 
by the shrinkage of the paint fihn. 


it will be noted that after reaching the maximum weight 
at " surface dry " it falls off in weight very slowly, while 
poppy and nut oil fall off in weight rapidly, which means 
a rapid shrinkage at this stage, a shrinkage which will 
cause cracking in the film itself or in a superposed film. 
In addition to these defects Professor Eibner has 
shown that owing to the low content of linolenic acid 



Fig. 16. 

the dried films of poppy and walnut oils are much more 
easily attacked by solvents. On the other hand, it 
must be remembered that all these vegetable oils, even 
olive oil, will dry in time, and it is, at any rate possible, 
that with time the poppy and walnut films will continue 
to improve. 

Against Professor Eibner's conclusions we have the 
fact that walnut oil was used from the earliest times, 
and that there is no suggestion by early writers of 



painting that they found walnut oil unreliable, and that 
there is also considerable evidence that the Dutch 
painters of the seventeenth century used poppy oil to 
grind with white lead and with the more delicately 
coloured pigments. 

It has also been shown by Dr. Morrell that under 
certain conditions linseed oil paint will crack freely. 

The most important conclusion to be drawn from 
Professor Eibner's and Dr. Morrell's researches is that, 
while it is safe to paint into wet oil (the method used 
by the Pre-Raphaelites), it is very dangerous to paint 
over a film just surface dry. When this stage has been 
reached the picture should not be continued for two or 
three weeks. 

Cracking is, of course, as much a question of careful 
manipulation as it is of the nature of the oil. Thin 
painting, with time to dry between each layer, is safe 
with any oil. Modern impasto puts an undue strain on 
a drying oil, and the marvel is that all the modern 
pictures do not crack — not that some do. As long as 
painters understand that there is increased danger of 
cracking in the use of poppy oil and nut oil, there is 
something to be said for grinding pigments in them, in 
spite of Professor Eibner's conclusions. Evidently 
further inquiry is necessary, more especially an examina- 
tion of the Dutch pictures, with a view to deciding how 
durable the films of these oils become in time. 

I have already stated the conclusions come to by 
Professor Eibner as to the yellowing of drying oils. It 
therefore remains to describe some experiments which 
throw further light on this question. 

The rate of yellowing is hastened by darkness and by 
the presence of moisture, and therefore in order to 
obtain results in a comparatively short time I enclosed 
the samples, painted out on glass slides, in the dark and 


By per. 

A WOMAN BATHING. By Rembrandt. 
(No. 54, National Gallery, London.) 

On the whole, Rembrandt's pictures are in good condition. Some of them are very low in 
tone, but this seems to be due to subsequent varnishing with a varnish which has yellowed 
badly, the paint below being little degraded from its original condition. The most durable 
variety of linseed oil, and the kind that yellows least, is stand oil which has, for centuries, 
been prepared in Holland, and Professor Eibner suggests that it may have been used by the 
early Flemish and Dutch painters. 



The sigiiificauce of the peculiar texture of the paint as suggesting the nature of the oil used 

is fullr dealt with in the text. Whatever the nature of the medium, this photograph of 

Rembrandt's brush-work should prove of great interest to the art student. The thick and 

sticky nature of stand oil. and the peculiar testiire of Rembrandts' brushwork. as revealed iu 

this magnified photograph, seem to make it possible that he mar have used this oil. 

For further information the reader is referred to Chapter XI. 


in an atmosphere saturated with water. A flake white 
and a zinc white used for artists' purposes were ground 
with various samples of oil and varnish, weighed quanti- 
ties of the pigment being ground with weighed quantities 
of the medium. The oils used were — a linseed oil, cold- 
pressed and sun-refined over water, a pure sample of 
poppy oil, pure walnut oil, which was also bleached and 
refined in the sun, a Dutch quick-drying Stand oil con- 
taining driers, a mastic varnish, copal varnish made by 
dissolving hard Sierra Leone copal in the specially pre- 
pared linseed oil and a dammar varnish made by 
dissolving 20 per cent of dammar in the Dutch Stand 
oil, and a carefully rectified specimen of American oil 
of turpentine. 

The samples were painted out on glass slides and 
allowed to dry in ordinary daylight, and then placed in 
the moist dark atmosphere under a bell jar. After one 
month they were examined and compared with fresh 
grindings of the same pigments and oils. Out of the 
twenty-four samples prepared, zinc white in poppy oil, 
white lead in walnut oil, white lead in an emulsion of 
Stand oil and yolk of egg, white lead ground in the 
Stand oil, dammar varnish thinned with a little turpen- 
tine, and white lead ground in a mixture half of linseed 
oil and half of mastic, had shown least change. 

The mixture of oil in mastic varnish proved unsatis- 
factory, the surface being easily powdered and brittle, 
so it must be excluded from these successful results. 
The superiority of walnut and poppy oil as compared 
with linseed oil was clearly shown, a sample of white 
lead ground stiff in the sun-refined linseed oil having 
become distinctly yellow. White lead ground in a 
double quantity of linseed oil had yellowed very much 
more than a white lead ground stiff, showing the impor- 
tance of keeping down the quantity of oil. 



A sample of white lead ground in oil thinned with 
turpentine had not yellowed any more than another 
sample ground in the same quantity of oil without 

These results are very much what we might have 
expected, but the behaviour of the Stand oil varnish 
is of special interest and also the Stand oil egg yolk 
emulsion. This emulsion does not crack on drying 
even if laid on thick, and is short and crisp under the 

The experiment with zinc white and poppy oil was 
very satisfactory ; but, as I have already stated. Pro- 
fessor Eibner has thrown great doubt on the value of 
poppy oil as a permanent medium, and it seemed 
advisable to compare flake white with zinc white in 
linseed oil. If both pigments are ground as stiffly as 
possible, zinc white requires about double the amount 
of oil that flake white requires, and 2 grammes of white 
lead with -3 grammes of linseed oil is of a convenient 
stiffness, but 2 grammes of zinc white require '6 grammes 
of linseed oil. 

If the flake white and the zinc white are ground with 
equal quantities of oil, the yellowing is equal, but if 
both are ground to an equal working stiffness the zinc 
white yellows more than the flake white, as it contains 
more oil. 

If, therefore, an artist wishes to use colours ground 
in linseed oil, he will find that stiff flake white will retain 
its whiteness better than stiff zinc white. Flake white, 
with equal amounts of copal oil varnish and linseed oil 
respectively, yellow to about the same extent. The 
bad reputation of copal varnish as a medium seems, 
therefore, to be due to its being added in excess of the 
amount of oil already present. 

An examination after six months proved poppy oil 



to change least, Stand oil without dryers and walnut 
oil to come next, and that dilution with turpentine had 
caused some yellowing, showing petroleum properly 
rectified to be the safer diluent. The copal oil varnish 
had yellowed a good deal more, but the Stand oil with 
dammar dissolved in it at a gentle heat had changed 
very little. The mediaeval varnishes made with soft 
resins probably therefore yellowed much less than hard 
resin oil varnishes of to-day. Dammar in Stand oil as 
an artists' varnish is worth the attention of artists' 

As it is known that the yellowing of the oil is due to 
the action of moisture, the question of protection of the 
oil from moisture becomes of importance. 

In a former chapter I have explained why, on account 
of the permeability of oil to moisture, it is advisable to 
protect the back of the canvas as well as the front of 
the picture. The importance of this is clearly brought 
out by the following experiment : 

Some flake white ground in linseed oil was rubbed 
out, A on a sheet of glass, B and C on ordinary artist's 
primed canvas. When the films of paint were dry, 
B was varnished with mastic, and C was both varnished 
with mastic in front and the canvas soaked with mastic 
at the back. Of these, B, the canvas plus mastic had 
yellowed most, the rubbing on glass came next, and the 
whitest was canvas C with mastic on both sides. Evi- 
dently the mastic had not proved so good a protective 
surface as the glass, and the primed canvas had proved 
of no value at all as a protective agent, while the white 
lead oil film sealed on both sides had remained practi- 
cally unchanged. A great deal of the darkening of our 
pictures is doubtless due to the neglect of protecting 
the back of the canvas by an impervious coating, so as 
to prevent the passage of moisture. 




In the last chapter, as a result of some simple optical 
experiments, we came to certain conclusions which have 
a very direct bearing on our methods of painting, and 
now require further consideration. Those results are 
not only a valuable guide in modern practice, but they 
throw an illuminating light on much of the early tech- 
nique in oil painting, and by studying this technique 
in the light of our experiments we shall gain much 
useful information. 

Let us then briefly restate the practical conclusions 
to which we have come. 

The object of the experiment was to determine, 
among other things, how to paint a picture in an oil 
medium which should remain high in key. 

It is obvious, if the pictures which have not been 
painted very long are examined, that there is a tendency 
for them to go down in tone. This varies very much 
with different painters. We have an infinite variety of 
technique among painters in oil, and the technique of 
some leads to a much more serious lowering of tone than 
the technique of others. 

On the other hand, whether we look at the fifteenth- 
century painters in oil, or at many of those of a later 



date, we do not find that the pictures have seriously 
altered through the centuries. It is evidently, there- 
fore, of the first importance that we should thoroughly 
understand this matter. Is it due to faulty chemistry 
— to impure or wrongly prepared oils and pigments, or 
is it due to a faulty method of using the materials ? 

A very striking instance of this lowering of tone was 
brought before me by Sir Arthur Cope. He had two 
sketches which he had made some fifteen years ago of a 
light bit of woodland, full of sunshine and wild flowers. 
The one sketch was in size and the other in oil. When 
first done they were as nearly as the medium would 
allow, of the same colour key. The picture in size was 
glowing and brilliant : the picture in oil was, when 
placed beside it, dark and dull. I examined the oil 
picture carefully under the microsope, and found that 
the surface had to some extent become ingrained with 
soot. This could not be removed without removing 
paint, and accounted for some of our trouble. The oil 
also had yellowed. 

After washing with saponin and water, I placed the 
picture for some months in a north window. This 
improved it considerably ; the oil bleaching to some 
extent, but it still remained far behind the size picture 
in brilliancy. 

That experience and many other similar ones have 
been haunting me for years, and I have felt that there 
remained something unexplained in the use of oil — some 
unexplored corner in the properties of oils and pigments, 
which would throw light on these matters. The optical 
inquiries, the results of which are given in a former 
chapter, have, as we shall see, very definite practical 
application. Let us then discuss what the conclusions 
are to which we have come. 

We have learnt to regard our pigments as transparent 



pieces of coloured glass, and to realize that their opacity 
and brilliancy will depend on the optical properties of 
the medium they are mixed with. If mixed with size 
or egg, we get very nearly their full brilliancy and 
opacity, and owing to the fact that these mediums do 
not alter appreciably in refractive index after they are 
dry, this brilliancy remains practically unaltered by 

A permanent and unalterable white is the gesso on 
which an old panel picture was painted, the mixture 
of size and plaster of Paris if non-absorbent forming an 
opaque white background. The exposed gesso surface 
of old pictures has got soiled and dirty, but protected 
gesso remains of a pure white. If then we choose to 
paint in size or egg our troubles are over, as then we can 
select — which we can easily do — pigments which are in 
themselves permanent. 

We have here the technique of the early tempera 
painters, and of the tempera painters of to-day — painting 
on the white gesso in a medium which gives each 
pigment nearly its full value and will not change. 

If we now proceed to grind our pigments in oil, we 
at once notice a change. They are all deeper in tone 
and more translucent. The new medium has a much 
higher refractive index than size or egg, and conse- 
quently our paints are appreciably nearer to their true 
tints unmixed with white light, revealing something of 
their transparency. 

If this were all, if the oil medium were as reliable and 
unchangeable as the egg or size medium, then we could 
work freely without fear with the new medium. But 
unfortunately it has two vices. In the first place it 
changes in refractive index with age. This begins from 
the moment the oil paint, painted out in a thin layer, 
has begun to dry. The dry paint is already not so 


By perjytission oj 



London. I 

(Xational Gallery, 

Eonmey was trained by an itinerant portrait painter, who liad, no douht, sound traditions, 
and he does not seem to have deviated from them. A limited palette, simple direct painting 
without foozling, and his pigments ground with just the right quantity of oil. seem to have 

been his secrets. 

His pictures are in good preservation, though somewhat cracked, especially in the whites, and 

have kept up to a high colour key, showing little degradation of tone. 


Showing the modelling of the face. Very rarely Eoniney used a little asphaltimi, and thi- of 

course, has always gone wrong. Occasionally he painted very thinlv on a twilled canvas with 

hardly any pruning. The magnified photograph of a part of this picture illustrates his brushwork. 

For farther information the reader is raferred to Chapter XI. 


opaque, not so brilliant, as the paint squeezed out of 
the tube. This change goes on slowly but remorselessly 
through the years, though, of course, at a diminishing 
rate. We know that the slow chemical changes which 
take place in the linseed oil film are not complete after 
four hundred years, that the film is still improving in 
toughness and insolubility. As the changes in the 
refractive index are doubtless due to these slow chemical 
changes, the change in the refractive index is probably 
also going on. 

The pigments then are growing deeper in tone and 
more translucent, and hence it is that we have penti- 
menti, the under-paintings ultimately showing through. 

The other change, which is also closely connected 
with the chemical change which results in the " drying " 
of the oil, is the yellowing of the oil. Unfortunately for 
the painter, the better the oil the worse the yellowing. 
The researches of Professor Eibner show conclusively 
that it is the valuable constituent which turns the oil 
into a tough elastic film, which also by another change 
causes the yellowing. Linseed oil yellows more than 
poppy or walnut oil, but it also produces a far more 
durable and insoluble film — and they all yellow. 

It does not require a profound optical knowledge to 
see that this yellowing will affect blue and green paints 
more than red and orange paints, and that it will affect 
most the more transparent paints where the light is 
plunging deep into the oil pigment layer. 

These matters we have already discussed. The 
yellowing of the oil will affect the blues more than the 
reds, and will darken and blacken the transparent 
cobalt blue much more than the comparatively opaque 
cerulean blue. 

The table on page 124 summarizes the results. 

The changes in the oil make us regard our whites — 



whether flake white or zinc white — with suspicion. 
We have found both of them to consist of transparent 
crystalHne particles. 

Not only will they yellow with the yellowing of the 
oil, but they will also grow more translucent, allowing 
dark under-painting to show through. 

As I have stated in Chapter IX, this is easily found 
experimentally. Paint on a canvas or a panel, black and 
white squares with size, cover the whole with size to 
make it non-absorbent. Then paint layer upon layer 
of zinc white or flake white over the black and white 
squares till they are completely covered and invisible. 
To hasten the action of time, place the finished panel 
in a window. In a few months the pattern of the black 
and white squares will be seen coming through. 

In the chapter dealing with colour we made certain 
other experiments which are of fundamental importance 
if we are to understand the early oil technique. 

We found that if a pigment had been painted out in 
egg or size and had sufficient medium to be non- 
absorbent, the painting of oil on it, if the oil was colour- 
less, did not appreciably affect its brilliancy ; and we 
also found that while a transparent paint like cobalt 
blue mixed with a highly refractive liquid, when in a 
thick layer would look almost black, that when spread 
thinly on a light under-surface it resumed at once its 
full brilliancy. 

Let us now consider the early oil techniques in the 
light of these experiments. 

In merely mentioning the words " oil technique " I 
am aware that I am entering on dangerous controversial 
ground. Frankly I am going to make certain bold 
assumptions, and so avoid a controversy as to whether 
the Van Eycks had or had not a secret medium and 
what the nature of that medium was, and the extent to 



which their pictures were not only begun, but finished 
in tempera or in a water emulsion. 

We shall take as our guide certain unfinished pictures 
and we shall assume, and the unfinished pictures seem 
to justify that assumption, that the foundations of the 
pictures are laid in tempera, whether egg or size or 
water emulsion, and are finished in an oil preparation, 
whether oil or oil varnish or blend of oil and egg, which 
being essentially an oil medium, with the properties of 
an oil medium, would yellow with time and changing 
in refractive index make the paints more translucent. 

Let us begin, in the light of what we have learned, to 
examine the unfinished picture by Jan Van Eyck at 

This picture is on a white gesso panel and has been 
completed as a drawing — completed, that is, in mono- 
chrome before the oil painting was bep^y/i. ^^-^^^ ^^ ^J;;, 
colour, the blue of the sky, has been abeady laid in. 
Now let us proceed to lay this oil colour on this drawing, 
and consider what will happen in course of time. Let 
us take, for example, the robe of the figure, and in order 
to select a pigment which will be most affected by the 
changes in the oil, let us suppose that we have glazed 
it with a transparent, or, at any rate, translucent blue. 
In time the oil will yellow, thus degenerating our blue ; 
but also the oil changing in refractive index will make 
our blue still more translucent, so that more white light 
will be shining through from the white gesso reflecting 
the light from below. The original scheme of light and 
shade of black and white, in permanent monochrome 
below, will correct the flattening effect of the yellowing 
of the oil. 

Moreover, we have seen that a pigment deepens in 
tone merely from the rise in the refractive index of the 
oil — a transparent blue in thick layer turning black- 

K 145 


but that the remedy for this is to paint thinly on a Hght 
ground. We are therefore actually utilizing one fault 
of the oil to correct the other — we are ensuring that 
increase of brilliancy ■will result from the rise in refractive 

Let us next consider painting where white lead has 
been used — let us say in the flesh paintings, which we 
may regard as white lead tinted with red. The yellowing 
of the oil will yellow, darken, and degrade the white 
lead, but the increasing translucency of the white lead 
will increase the luminosity from the white gesso below. 

I shall now proceed to examine another unfinished 
picture, attributed to Cima da Conegliano, in the 
National Gallery, Edinburgh. In this picture we have 
the opportunity of studying the method of oil painting 
at a later period. This painter is said to have been 
strongly influenced by Giovanni Bellini. The picture 
IS xii d, Liilf-finisliea c'^lMition, t^^*^ laying on of oil paint 
just begun. The whole has been painted first in a 
tempera monochrome, over which the oil paint is being 
laid, so that we have here the same technique, with 
little change, which we found in the unfinished Van 

The next unfinished picture which we shall examine 
shows us another aspect of this early technique. 

In the National Gallery, London, we have two 
unfinished pictures attributed to Michael Angelo. The 
first we shall consider, " The Descent from the Cross " 
(No. 790), was formerly described as a tempera picture ; 
but is now rightly described as an oil picture, or rather 
an example of the combined oil and tempera technique. 
We have here, again, the white gesso as our foundation 
white, and the preparation for the oil, in tempera 
monochrome ; but we have in addition an example of 
an under-painting in tempera with a bright colour. 


L":N'FIXiSHED PICTURE. By CniA da Conegliano. 
(Xational GiUery, Eiinburgh.) 
is picture is of great inteiest as revealing the oil tempera technique at the close of the 15th and 
,iiing of the 16th century. The whole picture has been laid in in a tempera monochrome on a 
•: gesso and the over painting with oil has just been begun. For the significance of the facts 
revealed by this unfinished picture the reader is referred to Chapter XI. 

y the Xatioiial Gallery 

MA1)0X>'A AND CHILD. By Michael Angelo. 
(National Gallery, London.) 

This unfinished picture in tempera oil technique by Michael Angelo is of special interest. On 
a white gesso panel the picture has, in the first place, been laid on in tempera. The virgin's 
robe, which doubtless he intended to glaze with real ultramarine or Azurro della Magna, has 
been completely modelled in black and the white of the gesso, while the fiesh tint.s have been 
deliberately lowered by a thin wash on the ges-so of terre verte. The significance of this 
technique is fully discussed in the text. 
For further information the reader is referred to Chapter XI. 


The figure on the left of the spectator, supporting the 
dead body of the Christ, wears a garment laid on in a 
brilliant red on the white gesso, either vermilion or red 
lead. The colour is very brilliant and fresh, and is 
ready to receive the subsequent glazings of translucent 
oil paint. 

For example, let us suppose a lake, in oil, laid over 
this red. The yellowing of the oil will degrade the lake, 
but the thin painting of the lake on the bright red will 
ensure no lowering of tone owing to the rise in refractive 
index of the oil. On the other hand, the rise in refractive 
index, by increasing the translucency of the oil paint, 
will ensure a more brilliant light from underneath. 

The next picture we shall consider is the other un- 
finished Michael Angelo (No. 809). 

Here, again, in the drawing of the Virgin's robe we 
see a completed black and white monochrome, the 
white being the gesso, ready for the glazings of trans- 
parent blue, probably Azurro della Magna. 

But this picture is of interest for another reason. 
Cennino Cennini, whose MS. we shall discuss fully later 
on, advises tempera painters to lay a thin coat of terre 
verte over the flesh, as a ground painting for the subse- 
quent flesh tints. It has been assumed by some that 
this direction of his indicates a universal practice on 
the part of tempera painters. In most cases their flesh 
painting is too high in key for this to be probable, and 
it certainly is not the practice in any oil pictures I have 

But whatever may be the truth of this matter, 
Michael Angelo has adopted the practice in this picture, 
the uncompleted figures having the flesh covered with 
a dull thin wash of green, and if we turn to the finished 
fif^ures, the low tone of the flesh as compared with the 
draperies is conspicuous. 



What is of interest to us is the proof of the value 
Michael Angelo laid on the pure white gesso under- 
lying his paint, since he deliberately stains it when he 
wants to lower the tone of his painting. That the green 
under-painting is supposed to give quality to the flesh 
tones is neither here nor there for our present purpose, 
which is the value in thin oil painting of the white 
gesso underneath. 

During the whole of this chapter we have made an 
assumption that the gesso itself was non-absorbent. 

On this matter Cennino Cennini's instructions are 
quite definite. 

Having described how to paint in oil, he says : "In 
the same manner we may work on iron, or stone, or any 
panel, always sizing first." 

It is evident that if the gesso is absorbent and stained 
with oil it loses all its optical value. When panel pictures 
have had the panel planed away the gesso is always 
found to be pure and white. Some have objected that 
if the gesso is non-absorbent the oil layer will not remain 
firmly attached. In my own experiments I have found 
that an oil film does remain firmly attached to a film 
either of size or albumen ; but evidently this fear was 
sometimes present, as is shown by the examination made 
by Mr. Thompson, of Harvard University, of an old 
panel picture. 

This panel had an interesting history. 

It was a picture of the Holy Family, attributed to 
Correggio. The panel was old and worm-eaten, the 
painting obviously late restorers' work. Mr. Brown, 
the picture restorer into whose possession the panel had 
come, determined to treat it with solvents in order to 
discover if there was any genuine early painting below. 
The result was that all the modern paint was removed, 
and the panel was found to have on it the outline of the 



This unfinished picture on a very old panel has an interesting history. Originally completely 

covered with paint it was attributed to Correggio, but was evidently pure picture faker's work. 

Mr. Brown, the picture restorer, removed all the comparatively modern paint, revealing the un- 

fliiished picture as shown in the photograph, probably late 16th century work. 

Mr. Thomson, of the Fog? Art Musemn. made a careful examination of this panel, and found that 

it was coited with a non-absorfcent white gesfo covered with a very thin skin of absorbent gesso 

into which the oil had soaked. He also found the oil painting of the high lights of the flesh 

very thinly painted compared with the shadows. The bearing of these results on the early oil 

technique is discussed in the text. 

Bv /•erii.issioii o/ -1/ or.-'. Macmil!iin ir ' 


By the Couitesyof Mrs. Hamilton Eruce. 

An exaiiilnatioii of the pigments of this picture showed it to be of the late 16th century or early 

17th century. A sample was taken through the dark background and through the high light of fht- 

flesh. The whole picture is on a white gesso lanel.and it was found that while the paint of the dark 

background was some three thousandth of an inch in thickness, the llesh painting was not more than 

three ten-thousandth of an inch, being the thinnest possible painting of white lead in oil on the white 

gesso below, thus ensuring for all time the high lights of the flesh. 

For farther information the reader is referred to Chapter XI. 


figures of the Virgin and the Christ and St. John painted 
on the white gesso in oil. Mr. Thompson subjected this 
panel to a very thorough investigation. The gesso was 
composed of whitening and size and was non-absorbent 
to oil. But on the surface of this gesso a very thin layer 
of plaster of Paris had been laid, only a small fraction 
of a millimetre thick. This layer was absorbent, and 
by cutting sections through the gesso and staining it he 
found that the oil had penetrated this very thin layer 
but had gone no further, thus on the one hand binding 
the oil to the gesso, on the other hand preserving the 
full luminosity of the white gesso below the oil painting. 

This panel also enabled us to test another matter in 
connection with the oil technique. 

Some years ago I had in my possession a portrait of 
a lady of about 1600, judging by the style, painted on 
an oak panel. I took borings through thedark painting 
and through the flesh painting and measured the thick- 
ness in each case of the paint layer. In the dark parts 
of the picture the thickness of the oil paint or the gesso 
was some 3 or 4 thousandths of an inch, about the usual 
thickness of a paint layer as laid on by a house-painter. 
But the flesh painting on the white gesso was only 2 or 3 
ten-thousandths of an inch in thickness, being the thinnest 
possible glazing of tinted white lead ove»the gesso below. 

Tests made by Mr. Thompson on the panel already 
described confirmed these results. 

The painting of the Infants was very thin and trans- 
lucent, the suggestions of outline in charcoal showing 
plainly through ; but the paint in the shading was much 
thicker than the paint in the high lights of the flesh. 

From its style I should place the panel as late in the 
sixteenth century. Both these pictures show a departure 
from the early technique. The picture is no longer first 
laid in in monochrome or in colour with a tempera 



medium. The oil is laid directly on the white gesso, but 
the value of the white gesso is still retained, as is shown 
by the very thin painting of the high lights of the flesh 
over the white ground. 

I shall take one more example. 

When the Rokeby Venus was slashed by a Suffragette, 
Sir Charles Holroyd gave me the opportunity of making 
a thorough examination of the picture and the detached 
pigments. The result of that examination and the proof 
of the authenticity of the picture has been published 
elsewhere. Here we are only interested in one result of 
the examination. 

The priming is of white lead, and would therefore 
degrade in tone to some extent compared with gesso, 
but still would retain much of its whiteness. 

On this priming a layer of deep red paint had been 
laid except under the flesh of the Venus. Here there is 
nothing but white upon white, and we are all familiar 
with the way in which the nude figure stands out lumin- 
ous against the dark background. Whether this was a 
habitual practice by Velasquez I do not know. I am 
inclined to doubt it, as, compared with his other pictures 
in the National Gallery, this nude figure is very vivid. 
It may have been an experiment. ^ 

If we turn from the examination of actual pictures to 
writers on painting, from Vasari onwards, we shall find 
no clear indication of a definite method based upon the 
optical properties of the medium, the priming, and the 
pigments. Van Mander tells us that Van Eyck covered 
his white gesso with a thin translucent, warm tone. It 
may be so. The painter having secured his white back- 
ground may play with it as he likes with translucent 

I have produced sufficient evidence to show that in 
the fifteenth century and for some time later there must 



have been in existence studio traditions for a correct use 
of a combined oil and tempera technique which were 
based on a complete understanding of the optical proper- 
ties of oil, though not, of course, an understanding in 
our sense. They knew nothing of the scientific principles 
involved, but by trial and error and experience there 
was in the time of Van Eyck and his followers a thorough 
knowledge of the use of this medium, in which its many 
defects are utilised to correct each other. 

When we come to later times the trail becomes con- 
fused. The writers on painting have no longer a grip 
on the fundamental optical principles, and to Rubens is 
ascribed the theory of solid lights and transparent 
shadows which has influenced so much the technique of 
the modern painter. 

Let us examine this also from the point of view of the 
optical properties of oil and pigments. 

In some of the Dutch painters, like Pieter de Hooch, 
we find especially in their panel pictures a triumphant 
permanency and brilliancy which equals that of the 
early Flemish painters. It is evident that their method 
closely resembles the old technique : the white ground 
is covered with a thin painting in oil. Under this thin 
painting there may have been translucent tints of the 
white priming. 

In his Practice of Oil Painting, Mr. Solomon J. Solomon 
says : — 

" With Terburg, JMetsu, de Keyser, Hals, and occa- 
sionally with Velazquez and many others, the grounds 
were prepared with varying tones of warm or cool grey 
evenly laid, graduating to a hghter ochre colour in the 
lower part, which is to represent the floor, and on this 
ground incidents, and accidents of hght and shade, are 
thinly indicated, such as the blacks and dark draperies 



generally. Where this preparation is not much darker 
than the half-tones of the scheme, the light of the picture 
does not suffer ; but where it is considerably darker, we 
find that the tonahty of a work is affected throughout, 
and most strikingly in the less solidly touched parts." 

The concluding words are significant. Sometimes they 
overdid the lowering of the white by a tliin wash of pig- 
ment, and he notes the result of the blunder. 

The result of the examination with the skilled, trained 
eye of an artist of these pictures is of much significance 
as supporting the conclusion, to which we have come, 
based upon purely optical grounds. 

But to return to the question of solid lights. Let us 
look at some examples. 

We can find one in a picture by Pieter de Hooch in 
the National Gallery, L. Xo. 832. The picture, which is 
on panel, has e^■idently been painted thinly on a white 
panel and is full of the luminosity to be expected ; 
but for his very highest light, where the light is falUng, 
for instance, on the sleeve of the man sitting by the 
table, he has put a solid thick painting of white lead. 

We find, then, two techniques combined in the picture, 
but in order to do this he has had to take obvious pre- 

The white lead is ground very stiff with the minimum 
of oil ; in fact, I suspect it of being merely mixed vdih 
the oil -snth the palette-knife, as it seems to contain 
unground particles. 

The whiteness of the white where used solidly by these 
Dutch painters is very remarkable and raises questions 
as to the oils they used. 

According to the authority of Willem Beurs, the Dutch 
painters used poppy oil for grinding their whites and 
their more delicately coloured pigments ; but modern 


' -• - (Xational Gallery, London.) 

Thi.s picture is an interesting example of Dutch technique. It is also of interest as being an 
example of peutemento, the tiles .showing through the woman's drei-s, owing to the very thin 
painting in oil having gradually grown more translucent with age. Beginning with a white 
grnunil, thin translucent oil paint has heen laid on the huninous non-absorbent white ground. 
Where a very high light is required, as on the sleeve of the man sitting at the table, solid 
white lead has been used, but stiffly and coaisely ground in the minimum of oil. The white- 
ness of the whites in the Dutch pictures is remarkable. 
The (iue>tion of the oil, whether Poppy or Dutch Stand oil, has not yet been settled. For 
further information the reader is referred to Chapter XI. 

Ey permiiiioH vj ''" ' rushes o/ llu- 

(National Gallery, London.) 
To Rubens is attributed the advice to paint the high lights solid and the shadows translucent, 
a reversal of the methods of the earlier painteisin oil. 
In this picture, however, on a white gesso panel, he has followed very closely the best traditions 
of the Dutch School, based on the Flemish School of Van Eyck and his followers, and has 
painted thinly on the white gesso ground. The picture is in excellent preservation and comraus 
favourably with many of his larger canvases from this point of view. 
For further information the reader is referred to Copter XI. 


research has thrown doubt on the rehabihty of poppy 
oil, and has directed attention to the unchangeable 
character of Dutch Stand oil. It is somewhat dark, 
sticky, and flowing, an abominable medium, but could 
be used either mixed with yolk of egg or thinned with 
turpentine, and yellows very slightly with age. 

These are questions worthy of further enquiry, but at 
present we are dealing with general principles. 

If a high light is to be produced by white lead and not 
by light from the gesso surface, the white lead must be 
ground very stiffly with a minimum of oil and either 
painted over white or painted very thickly. This pre- 
caution was understood by the early painters. Examples 
of the use of white for a high light sohdly painted is to 
be found in fifteenth-century tempera pictures. In the 
picture on panel by Rubens in the National Galler}' the 
portrait of Susanne Tourment (No. 852), which is in fine 
condition, we see that he could appreciate the value of 
tliin painting on a white gesso. In his large oil paintings 
we often find yello%nng and degradation of tone, the 
boldness of the design and of the colour scheme hiding 
these defects. It is not without significance that, accord- 
ing to contemporary records, Rubens only used oil of 
turpentine as a diluent, thus keeping the oil content as 
low as possible. 

Thousands of pictures show in the masses of brownish 
yellow white lead they contain the danger of the theory 
of solid lights when applied by men who had no under- 
standing of the optical properties of oil. 

The interesting question arises here of how thick and 
stiff white lead must be to be safely painted over a dark 

In my experiments I have found that black and white 
squares would begin to show through a stiff layer of 
white lead in oil of 2 thousandths of an inch in thickness. 



In the Van Dyck, No. 2127, in the London National 
Gallery, the lace of the collar has been painted in white 
over black, but very stiff paint has been used, and it has 
been put on in thick blobs. In No. 1399, " Portrait of 
a Gentleman," by Gerard Terborch, one of the most 
perfectly preserved pictures in the Gallery, solid white 
has been used on the high lights, and is very bright and 
pure, and seems in places to be painted over the dark 
coat below. 

Another interesting example of the painting of white 
over black is to be seen at the side of the window in 
Rembrandt's " Philosopher " (No. 3214), the white 
having been laid on over black and then removed, 
apparently with the end of the brush handle, showing 
the black below. On the whole, his paint is thick enough 
to stand it, but it is not so luminous as the white of the 
window, laid thickly over a light-coloured ground. In 
this and many other pictures Rembrandt's brushwork 
suggests the use of Stand oil. 

Another interesting question about the Dutch painters 
is whether they painted their canvas pictures on a gesso 
ground or on an oil ground. They seem to have painted 
indifferently on panel and on canvas, and on the whole 
their panel pictures are the best preserved ; but their 
canvas pictures are remarkably good and of a high 
luminosity. The ground they painted on, therefore, is 
of the greatest practical importance. 

I have, I think, said sufficient to show that Van Eyck 
and his followers had a complete understanding of the 
optical properties of their pigments and media and made 
full use of that knowledge ; and that later on, as new 
methods of painting were evolved, the more careful 
painters still followed the old tradition and utiHzed new 
methods with caution and common sense. They were 
jiilly aware of the follo^nng facts : — 


the Truslc-ts of the 

(National Gallery, London.) 

One of the subjects discussed in the text is the gradual increase in translucency of white lead 

ground in oil. 

This picture is of interest because Van Dyck. painting at a time when the technique of oil 

painting was thoroughly understood, has deliberately, in painting the lace of the collar, laid 

white over black. The white, very stiffly ground, has been laid on in little thick blobs, 

thick enough and dense enough to keep up their full tone of white through the centuries. 

For further information the reader is referred to Chapter XI 

By permission oj '•:; Trusttrs o/ f/w Xatiottat Cw.iilery ^ 

(National Gallery, London.) 

This picture has been selected to show how far it is safe to paint white over black. The lower 

toned white along the edge of the illuminated wall is painted over black, and here and there 

the stiff white wet paint seems to have been removed with the wooden end of the brush. 

The main illuminated surface is white laid very stiffly on a greyish toned background. 

For further information the reader is referred to Chapter XI. 


1. That oil yellows vdth time. 

2. That pigments in oil get more translucent with 


3. That pigments in tempera are much brighter than 

pigments in oil and retain their brightness. 

4. That glazing in oil over pigments in tempera does 

not impair their brightness if the tempera surface 
is non-absorbent. 

5. That transparent pigments suffer more degrade tion 

in oil than opaque pigments, and should there- 
fore be used for glazing and scumbling on solid 

6. That this is more especially true of blues and 

greens. See the black and white modelling of 
the Blue Robe, already referred to, and the 
practice of Venetian painters of putting on blue 
mth size. 

7. That a pigment which suffers degradation in tone 

from the rise in the refractive index of the oil 
has its brilliancy fully preserved if pai.ited thinly 
on a bright under surface. 

8. That the most brilliant permanent white is the 

white gesso of the panel, which should therefore 
be utilized wherever possible. 

9. That solid whites in oil -wdll go down in tone unless 

the minimum of medium was used, the paint 
being put on very dry and, if there is dark paint- 
ing below, very thick. 




In Chapter V we have dealt fully with canvas and panel 
and the priming of both with oil paint or gesso. We 
have learned there that the best foundation for a picture 
is a panel of old seasoned mahogany, or 5-ply well- 
seasoned birch, painted with oil paint and with canvas 
glued on, which is then coated with gesso. Instructions 
are also given for priming a canvas with oil. 

An oil primed canvas, whether prepared by the artist 
or purchased, should be kept at least six months before 
painting on it. During this time the oil will have had 
time to dry, and the canvas will probably have darkened 
and will represent more nearly the tone which it will 
ultimately take. A white canvas is preferable to a grey 
canvas. It is better for the artist to tone down the 
surface for himself with a thin transparent wash, and 
still better to block out the future high lights and shadows 
in a neutral tint. 

If a very thin layer of oil is painted out on a sheet of 
glass and weighed from time to time, we can follow the 
process of the drying of an oil. The oil absorbs oxygen 
and moisture from the air, thus increasing in weight, 
but at the same time, owing to the complex changes 
going on during the oxidation of the film, it is losing 
certain volatile products formed during oxidation. 
During the first stage of the drying it is rapidly in- 
creasing in weight, and during the latter stages losing 
weight and shrinking. It goes through four stages. It 


Jly ft-rniiisioH y the Trustees of the Sational Gallery 

PORTRAIT OF A GENTLEilAK. By Gerard Tebborch (1617-1681). 

(N'ational Gallery, London.) 

This picture is in perfect condition, and the whites are pure white and have not yellowed 

with time. As it is known that white lead in oil gets more translucent in time, this picture 

raises interesting questions as to how far the white overlies the black, and how far it is on 

a white priming on the panel. 

For further information the reader is referred to Chapter XI. 


is first wet, then sticky, then surface dry, and finally 
dried through. Drying through takes a long time. In 
fact,' slow chemical changes are still going on after four 
hundred years, and an oil cannot be regarded as dried 
through under a few weeks. During the change from 
wet to sticky the oil is not contracting, the most rapid 
contraction taking place just after it is surface dry, and 
then after that the rate of contraction getting slower 
and slower. 

The time, therefore, of danger, when a fresh layer of 
oil is apt to be cracked, is when the oil film below has 
just reached the stage of surface dry. It is quite safe 
to paint into wet oil ; but once the oil is surface dry 
it should be given some weeks before paint is laid 
above it. Linseed oil loses much less weight than 
poppy or nut oil, and therefore is not nearly so likely 
to crack. It may possibly be better, therefore, to give 
up the practice of grinding colours in poppy oil, though 
it has the advantage of not yellowing so much as linseed 
oil. The first layer of paint should contain a higher 
percentage of pigment to oil than the next layer. 

Thin painting is much less likely to crack than thick, 
but the main thing is to allow a sufficient time for drying 
between each layer of paint, and, if possible, the dried 
surface should be rubbed down between each painting 
with fine sand-paper or pumice. 

Volatile mediums containing resins dissolved in them 
should be avoided. They are apt to cause cracking. 
In fact all methods of hastening the painting of the 
next layer over the layer below are dangerous. 

In the chapter on optics we have fully described the 
effects of yellowing and the change in refractive index. 
We have now to apply what we learned in those chapters 
in practice. 

The pigments supplied ground in oil by the artists' 



colourmen contain quite sufficient oil, and in the case 
of \dridian, cobalt blue, madder and alizarin lakes, and 
French ultramarine the oil is apt to be in excess, and 
will float up and ultimately darken the pigments. Only 
stiff white lead should be used. The thin varieties 
contain an excess of oil. 

If possible oil, varnish, and oily mediums should be 
avoided. This merely means increasing the oil content 
and therefore increasing the ultimate lowering of tone 
of the picture. No volatile medium shoidd be used 
containing soft resins ; they are apt to cause cracking. 
The diluents should either be well-rectified turpentine 
or better well-rectified petroleum ; neither should leave 
the slightest stain on blotting-paper soaked in them 
and left to evaporate in the air. 

The artist should paint out samples of the colours 
which he uses pure from the tube, and diluted with 
different quantities of turpentine. It is easy to see if 
any of them have an excess of oil. In the case of these 
some of the oil should be removed before using with 
blotting-paper, care being taken not to remove too much. 
Some of the colour, after mixing with turpentine, may 
dry dead. This does not mean that they have not a 
sufficiency of oil to bind. To test this, after they are 
dry, rub with a rag placed over the finger and moistened. 
If no colour comes off they are bound. To remove the 
matt or dead appearance, rub on a little oil with the 
finger and wipe off with a rag. This will probably be 
found quite sufficient, and if not, the surface can be 
gently polished with a soft rag when dry. If the colour 
can be rubbed off, the treatment with oil vnll be found 

One pigment if used pure requires special care, and 
that is cadmium yellow. In course of time the cadmium 
yellows act on the oil so that they can be washed off. 




till- Trnsteis of llie A'alional Gallery of British Art 

MR. AND MRS. LIN'DOW. By Rosixey. 

(Tate Gallery, London.) 
Koniney's method of painting has been fully dealt with in the description of the " Lady and child." 
The magnified photograph of a portion of this picture showing its brushwork is of great interest. 



A little copal varnish, either rubbed on or mixed with 
the paint, is a wise precaution in the case of these yellows. 
Generally there is no objection to the use of copal varnish 
instead of oil if used in this cautious way. If a picture 
has dried dead in parts, oiling out should be done in the 
way already described, all excess of oil being wiped off, 
and if still not shiny enough, try polishing with a silk 
liandkerchicf before resorting to more oil. If a colour 
is oily, a thick layer is worse than a thin one, as more 
oil rises to the surface of a thick layer, and will therefore 
when it yellows cause more lowering of tone. 

It is safer to dilute with turpentine and rub on a little 
oil in the way described above than to dilute a colour 
with oil, as too much oil is apt to be added. If a painter 
finds the addition of some oil necessary to get his results, 
a little oil, or oil and copal varnish thinned with tur- 
pentine is safer than oil alone, as he is less likely to add 
the oil in excess. Rubbing a dried oil sm'face with sand- 
paper or pumice before painting on it removes excess of 
oil which has floated up, gives a tooth for the binding 
of the next layer, and makes tlie surface very slightly 

It is not a bad plan, in the case of an oily pigment, 
to remove a little of the oil with blotting-paper, care 
being taken that too much oil is not removed so that the 
pigment will no longer be bound. 

Some artists like to add a little copal varnish to their 
pigments, thus following a traditional practice. If the 
copal varnish is added to the pigment as supplied ground 
by the artists' colourmen, there vnll be an excess of 
medium and the picture will go down in tone. Copal 
varnish must be used to replace oil, and therefore a little 
of the oil should be removed by blotting-paper before 
the copal varnish is added. Copal varnish yellows more 
than linseed oil. 



Owing to the present practice of supplying pigments 
ground in oil put up in tubes and to the demand on the 
part of the artists for a pigment in perfect condition, no 
matter how long it has been in the tube, buttery and 
crisp and short under the brush, artists' colourmen are 
compelled to add some substance of a transparent nature 
to certain pigments which Mill float the pigment in the 
oil. This tends to cause an excess of oil to be present 
beyond that necessary to carry the pigment, which prob- 
ably tends to lowering of tone. If artists would return 
to the practice of grinding the pigments in oil as they 
want them, probably much of the lowering of tone 
complained of would disappear. 

The pigments should be ground with a muUer on a 
glass slab, preferably with linseed oil cold-pressed and 
sun-refined, sufficiently stiff to stand up. Two or three 
drops of copal varnish can be added to each grinding. 
They can be kept in little jars ^^'ith a piece of oil 
paper or bladder over them. The white lead must be 
ground as stiffly as possible and can be kept under 

When we come to the appUcation of the results of our 
optical investigations in the properties of oil and pig- 
ments in oil painting we are in a difficulty. Every artist 
has his ovm technique which he uses to express his own 
personality, and naturally regards with impatience any 
restrictions on his methods. The painters of the fifteenth 
and sixteenth centuries were apparently %nning to accept 
certain definite methods for the building up of 
the picture and to express themselves through those 
methods. This is no longer possible or desirable, and 
therefore all that we can do is to lay down certain 
general principles. 

In the first place, a picture should start from as white 
a priming as possible. If painted on panel this can be 


(Wallace CoUectlou.) 

The magnified photograph of the brush work of this picture is reproduced with the view to 
showing the peculiar texture of the paint, which it is suggested mar have been a pigment 

ground in stand oil. 


BY EEIIBKANT. (Wallace Collection.) 

Quite apart from the interest of this photograph as revealing Rembrandt's brushwork. the 

peculiar texture bears out what has already bten said about the "Woman Bathing" as to 

the peculiar quality of his medium. 

Bol and others, while under Rembrandt's influence, produced a similar surface, but when 

niagni-ied photographs are compared they are extraordinarily weak and meaningless beside 

the work of the master. 


gesso, the most permanent and reliable of white ; if 
painted on canvas only an oil priming is safe, and there- 
fore a certain degradation of tone is inevitable, but must 
be reduced to a minimum. 

Beginning -vvith the foundation of pure white, if we 
imagine a section cut through the finished oil painting 
the gradation should always be from dark to light, 
starting from the outer surface to the priming below, 
and there is only one exception to this, namely, that 
flake white can be used for a high light if painted very 
solidly and stiffly, even though there is darker paint 
below ; but it is vnser even here, if possible, to avoid 
such darker painting. 

If an artist deliberately decides to put darker paint 
under lighter paint — as in the case already discussed, 
where Michael Angelo has put a dull green under his 
flesh — he must do so deliberately with knoAvledge that 
in time the effect of this darker painting is going to be 

Starting with the restricted palette based on purely 
chemical grounds, it should consist of the follo\nng 
pigments : — 

White lead, *zinc white, *permanent white, *yellow 
and *red ochres, *Venetian red, *Indian red, *cadmium 
scarlet, vermilion, cadmium mid, cadmium yellow pale, 
made by a furnace process, *barium chromate, aureolin, 
*cobalt green, *oxide of chromium, *viridian, *cobalt 
blue, ultramarine, Prussian blue, *permanent violet, 
*cobalt violet, madder and alizarine lakes. 

The Prussian blue is the most doubtful pigment in 
this list, and it is still an undecided question whether a 
carefully prepared and washed lead chrome should not 
be added. 

Those marked with an asterisk are the most permanent 
of all. 

L 161 


Of these pigments, a further selection is to be made of 
Venetian and Indian red, vermilion, opaque bright 
yellow ochre, cadmium mid and cadmium pale (made 
by a furnace process), cobalt green, and oxide of chro- 
mium green and cerulean blue for the bright solid 
pigments which will be least affected by alterations 
from the oil. 

These pigments should be used for solid under- 
painting, and should also be used where pure spots of 
colour are required on the picture. 

The madders, raw and burnt sienna, aureolin, viridian, 
cobalt blue, ultramarine, and Prussian blue should be 
used for thin glazings and scumbhngs and for tinting 
white. Lemon chrome occupies an intermediate position. 

If the artists' technique lends itself to thin painting 
on a pure white ground, after the manner of fifteenth- 
century painters and some of the Dutch painters and 
the Pre-Raphaehtes, these translucent pigments will be 
the most suitable he can use if he wishes to get the full 
value of the white reflected from the white ground below. 

Never forget that increased translucency is inevitable 
and will lower the tone unless we have always been 
careful to build up from light to dark. Dark under- 
painting is sure sooner or later to show through and 
lower the tone of thinner painting above. Stiff white, 
thinned if necessary with turpentine, should be used for 
high hghts, and, if there is dark paint below, should be 
laid on thickly. 

The less oil in the white, the better it will keep up 
its tone. 

To select the pigments which best keep up their tone, 
get whites and bright colours from different artists' 
colourmen, paint them out side by side on a canvas. 
When dry, place a pail two-thirds full of water in a 
black, dark cupboard, place the canvas on the top of 



the pail, lock up the cupboard, and examine for change 
of tint every month, filling up the pail when necessary. 

Finally, the picture should be varnished with mastic 
not sooner than six months after its completion, and 
during that time should be protected from dust and dirt. 
Mastic varnish is better than an oil varnish, because it 
not only protects the painting below from the attacks 
of oxygen and moisture and deleterious gases, but it has 
the advantage that it can be easily removed when it 
yellows and cracks without any injury to the paint film 
below. It should be laid on very slightly warm, the 
picture being gently warmed and dry, and rubbed over 
with a warm silk handkerchief before laying on the 
varnish. If care is not taken to do this, mastic is apt 
to bloom. 

The back of the canvas should be covered with tinfoil 
or with beeswax and resin as described in Chapter V. 

In conclusion, there can be no doubt that the most 
brilliant permanent white is gesso ; and that the com- 
bined tempera and oil technique gives pictures which 
keep up their tone better than any picture painted 
entirely in oil. 




The balsams and oleo resins exude from certain trees 
and are emulsions of resins, in oil esters in the case of 
the balsams, and in essential oils in the case of the oleo 
resins, but the word balsam is very often applied to both. 

Copaiba Balsam. Copaiba balsam is an example of 
a true balsam containing 40 per cent of oil esters and 
some 60 per cent of resin, and is the exudation of different 
varieties of copaiferse occurring in South America, East 
Indies, and Africa. I have mentioned its use in the 
cleaning of pictures from old varnish. It was used by 
Pettenkofer as a varnish, whose methods of cleaning 
pictures I have referred to, and is still used as a varnish 
by certain picture restorers in place of mastic. Its use 
for this purpose is worthy of further investigation. 

Strasburg and Venice Turpentine. Three oleo 
resins are of interest to the artist — namely, Strasburg 
turpentine, which the Italians called olio dAbezzo, 
which is obtained fiom the silver fir, and has some 
80 per cent of resin acids and resins dissolved in some 
20 per cent of terpenes, and which was used by the old 
painters as a varnish by thinning vnth a little oil of 
turpentine and also as an ingredient in their oil var- 
nishes, being sometimes used to dissolve the more 
insoluble resins like amber. Thinned with turpentine, 
it dries slowly and remains sticky for a long time, but 



ultimately makes an excellent varnish for the surface of 
a picture, and it is also at any time easily removed. 

Another is Venice turpentine, from the larch, con- 
taining about 80 per cent of resins and resinous acids 
and 20 per cent of terpenes. It was also used in the 
making of varnishes, as can be seen in the old recipes. 
Neither of these substances is used in varnish -making 

The third is Canada balsam, from the Pinus balsamea, 
which is the easiest of these balsams to obtain pure and 
reliable at the present day, and which, for those who 
wish to experiment with these materials, will be found an 
excellent substitute for Strasburg or Venice turpentine. 

All these balsams have a high protective value, when 
they have dried, as varnishes, resisting the entrance of 
air, moisture, and deleterious gases ; and I have already 
described in Chapter III how a transparent and per- 
manent green can be prepared from verdigris by dis- 
solving it in them. 

The question as to how far they entered into the 
medium used by the early oil painters has not yet been 
settled. A varnish made by combining them with oil is 
very soft and sticky, but hardens in course of time, and 
it may well be that, though they would be condemned 
by the modern varnish-maker as gi\nng soft and perish- 
able varnishes, they continue steadily to improve 
with age. 

Spirits of Turpentine. From these oleo resins, by 
the process of distillation, oil or spirits of turpentine is 
obtained. The American turpentine, which is chiefly 
used, is prepared from the oleo resins of two trees, Pinus 
australis and Pinus tceda. The turpentines are mixtures 
of bodies known as terpenes, the proportion of absorbent 
terpenes present depending upon the tree from which 
the turpentine has been prepared. All these terpenes, 



on exposure to air, oxidize, forming sticky resins which 
are very objectionable if introduced into the fihii of oil. 

We should therefore select for artists' purposes a 
turpentine which does not oxidize too readily. Sir 
Arthur Church regards the French oil of turpentine, 
Pinus maritima, as the least oxidizable, and next to that 
the American oil of turpentine. Owing to this tendency 
to oxidize, the artist should always have fresh turpentine 
and should keep it closely corked up so as to prevent 
oxidation. A simple test of its condition is to pour a 
little on blotting-paper and allow it to evaporate. It 
should evaporate clean, lea\ no oily stain. Sir 
Arthur Church advises putting a lump of quicklime in 
each bottle. 

Some artists are prejudiced against spirits of turpen- 
tine as a diluent, but it has been used so long and so 
universally that, if fresh and properly rectified, there is 
no reason to suppose that it can have any injurious 
effects. It is added merely to thin the paint, evaporating 
clean, and leaving the oil and pigments behind ; but it 
also acts as a drier, stimulating the drying of the linseed 
oil. If a large quantity is introduced, the painted surface 
dries flat or dead. It has a slight tendency to make the 
oil film turn yellow. 

Oil of Spike Lavender. Some artists prefer to use 
oil of spike lavender, obtained by the distillation from 
the flowers of a species of lavender known as Lavandula 
spica. The same precautions have to be taken with oil 
of lavender as \\ith turpentine, as it also oxidizes and 
thickens when exposed to the air. 

Resins and Varnishes. If the oleo resins are heated 
so as to distil off the terpenes they contain, we have the 
resins left behind. In the case of the ordinary turpen- 
tines, the resin left behind is known as rosin, or colophony. 
This resin, if dissolved either in turpentine or in linseed 



oil, forms a very soft varnish, and, while formerly used 
for this purpose, is not used in modern varnish-making 
where the oil is linseed oil, but, on the other hand, it is 
used for making varnishes with tung oil, for which it is 
excellently suited. There are a large number of resins 
coming from various sources and used in varnish- 
making, some of which are as follows : — 

Mastic. This resin is obtained from the Pistacai 
leniiscus, which is found in the Levant and the Greek 
Archipelago. Its specific gra\'ity is between 1-056 and 
1-060 and melts between 221° and 248° F. It is soluble 
in turpentine and alcohol and in oil. The picture varnish 
used by artists is made by dissolving mastic in tur- 

Sir Arthur Church gives the following recipe : — 

Dissolve 14 parts of ground mastic mixed with 6 parts 
of fine sand and 44 parts of spirits of turpentine by 
heating in a water-bath. Pour off into a bottle, cork, 
and leave to stand until it becomes perfectly clear. In 
order to make the varnish less brittle, a little Venice 
turpentine, Canada balsam, or Elemi resin, or a very 
little linseed oil may be added. The advantage of mastic 
as a picture varnish is the ease with which it can be 
removed without injury to the painted surface below by 
friction or solvents. 

Sandarach. This resin is obtained from Callitris 
quadrivalvis, a dwarf tree which grows in the north-west 
of Africa. It melts at 135° C, and is completely soluble 
in ether and hot pure alcohol and partially soluble in 
oil of turpentine. The mediaeval sandarach may have 
been this resin, but was more probably the resin of the 
juniper. It is used for making spirit varnishes. 

Dammar. This resin is obtained principally from the 
Dutch East Indies and British Malaya. It is the product 
of the Dammar a orientalis. It melts between 260° and 



300° F. It is soluble in turpentine, petroleum spirit, and 
in oil, and is used for making spirit varnishes. 

Lac. Lac comes principally from India. It is 
obtained from several trees, and is due to the action 
on the juices of the plant by the female insect Coccus 
lacca. This insect secretes the lac round its eggs, and 
finally becomes itself embedded. The lac as obtained 
from the trees contains a dye, o\\ing to the presence of 
the bodies of the insects. This dye is extracted by water 
and was formerly used in Europe, both for dyeing and 
for making lakes. The lac after the extraction of the 
dye is melted and filtered through canvas bags. Shellac 
is partially soluble in alcohol, the solution being what is 
known as shellac varnish. 

Copal. The name copal should be restricted to 
certain fossil resins which come principally from Africa, 
but also from Australia and other places. Unfortu- 
nately, two resins may be obtained from the same 
source, a fossil resin found some 3 or 4 ft. under the 
ground, and a resin from the same tree collected from 
the tree itself, which is much softer. For example, at 
one time Sierra Leone copal was one of the hardest of 
the fossil resins ; to-day Sierra Leone copal is collected 
from the living tree. The hard fossil resins to-day are 
Congo copal from Central Africa, and Zanzibar copal 
(or animi) from East Africa, angola from West Africa, 
manilla from Manilla and the East Indies, and kauri 
from New Zealand. The soft resins are also obtained 
from the same sources. Manilla copal, for instance, may 
be a hard fossil resin or a soft resin from the living tree. 
Only the finest fossil copals should be used for making 
copal varnish for artists' use. 

Amber. Amber is the hardest and most infusible 
of the resins. If fused and blended ^\ith oil, a very 
dark slow-drying dichroic varnish is obtained. Certain 



solvents, such as chloroform and epiclilorohydine, dissolve 
it partially, and it can be dissolved under pressure and 
high temperature in sealed tubes in turpentine and other 
solvents. The principal sources of supply are the 
Prussian shores of the Baltic Sea, where it is found in 
veins and quarries. It has a specific gravity of 1-07 and 
a hardness of 2-5, and fuses at about 300° C. It is 
practically insoluble in ordinary resin solvents. 

It consists principally of a resin called succinite, and 
on destructive distillation yields succinic acid. It is 
highly improbable that pure amber varnish was used by 
the early painters. It is more probable that the name 
amber varnish was applied to the varnish because of its 
colour resembling amber. 

Varnish. The process of making an oil varnish is 
the same to-day as it was in the eleventh century. The 
resin or resins are fused, the linseed oil is heated, the 
two are blended and heated together until a thorough 
union has taken place, and a drop of the varnish, when 
put on a glass plate to cool, remains clear and trans- 
parent. In the case of the modern varnish, driers are 
incorporated, and the whole is thinned down \vith spirits 
of turpentine. In the case of the mediaeval varnishes, 
it is evident from the recipe that they had no turpen- 
tine, and the varnish must have been very thick and 
sticky and had to be rubbed on hot, or required, if used 
as a painting medium, to be emulsified with yolk of egg. 
It is not till we come to the recipe for varnish-making 
of the sixteenth century that we find both alcohol and 
turpentine freely mentioned. As far as we can judge, 
the old varnishes contained a much higher percentage 
of resin to oil than the modern varnishes. Much softer 
resins, such as pine balsam, rosin, and sandarach, being 
used, so that they would not be as durable as a good 
modern copal varnish, but, owing to the high resin 



content, would form a better protective surface for 
fugitive pigments, and would not yellow so much. 

The softer resins can be dissolved, some in alcohol and 
some in turpentine, forming spirit varnishes. I have 
already given a recipe for mastic varnish. 

The distinction between the two types of varnish — 
the oil varnish and spirit varnish — is obvious. When 
the turpentine has evaporated from mastic varnish we 
have a thin film of the pure resin left which is practically 
impervious to moisture, but is brittle and fragile. 

In the case of the oil-varnish, the film which is left is 
a blend of oil and resin, and the drying of the oil varnish 
depends upon the drying — ^that is, the oxidation of the 
oil it contains. The result is a hard but flexible film of 
great durability and very insoluble, but it is not so 
waterproof as the film of pure resin, both moisture and 
gases being able to permeate the linseed-oil film. 

For the preserving of fugitive pigments the pure resin 
itself is best, and the method which was used by the 
old-fashioned coach-painter was to protect his pigment 
by mixing it with turpentine and resin, and then protect 
the friable resin by a coat of oil varnish on the top. By 
this method he obtained a very durable combination ; 
while, owing to the fact that the coach was constantly 
exposed to light, the yellomng of the oil varnish was not 
a serious objection. 

Another method of making oil varnishes is to begin 
by dissolving the resin in a powerful solvent like chloro- 
form, by boiling the powdered resin in chloroform for 
some time, the flask being fitted with an inverted 
condenser. Along with the chloroform a certain amount 
of turpentine is also added. At the end of the operation 
the chloroform is distilled off, leaving the turpentine. 
The dissolved resin is again boiled with the turpentine 
alone and then mixed with a linseed drying oil which 


Thie photograph reveals the fact that mastic varuish is farfrom a perfect protecting agent 
for the oil paint below. It seems to crack owing to movements in the oil film and the slow 
desiccation of the varnish. The cracks are quite different in character from those occurring 
in oil paint. What is required is a varnish which will protect the oil paint below from 
injurious gas or moisture as perfectly as mastic, and at the same time will not yellow 

or crack. 


has been heated, and the whole heated and stirred for 
some time. Amber can be partially dissolved by this 

For full and detailed information about resins and 
varnishes the reader is referred to Varnishes and their 
Components, by Dr. Morrell. 

Petroleum. Rectification of the natural petroleum oils 
which are mixtures of hydrocarbons are separated into 
oils of different boiling-points. The portion boiling under 
170° C. is called petroleum spirit and can again be 
rectified into oils of various boiling-points, the lowest 
being about 50° C. As a diluent for artists' use, accord- 
ing to Sir Arthur Church, no fraction having a boiling- 
point higher than 170° C. should be used. The petroleum 
should evaporate clean. Blotting-paper soaked in it 
and left in the air at ordinary temperatures should 
ultimately leave no greasy stain or smell. 

Beeswax and Paraffin Wax 

Beeswax. Reference has been made more than once 
to the use of beeswax for various purposes, both old and 
new. It is a fairly permanent substance, but not so 
permanent as seems to have been assumed by those who 
tried to reintroduce wax painting, as by long-continued 
exposure it disintegrates and partially perishes by oxida- 
tion. The excellent state of preservation of the pictures 
from Hawara has led to false conclusions. Not only 
were certain repairs necessary on them by remelting the 
wax, but it also must be remembered that they had been 
preserved in Egyptian sand. 

If beeswax is dissolved in turpentine, spread over a 
surface thinly, and polished, it forms a protective coat- 
ing against dust and dirt, but it has not a high value 
as a protection against the action of air, moisture, and 



injurious acids. From tests which I have made, I put 
its protective value at about one-tenth the protective 
value of mastic varnish. On the other hand, a layer of 
melted wax and resin, say about the twentieth of an 
inch in thickness, resists the attack of air, moisture, and 
acid fumes about ten times as long as mastic varnish, 
and may be regarded as practically indestructible when 
laid between layers of canvas. And it is for this reason 
that I have advocated it for the backing of canvas 
pictures in oil and tempera. 

Paraffin Wax. Paraffin wax, of vrhich the one with 
the highest melting-point is known as ceresin, is a modern 
product obtained from native petroleum and ozokerite 
and consists of a mixture of the substances known to 
chemists as hydrocarbons. These are kno^vn to be very 
stable and permanent substances, and not to be attacked 
by most chemical reagents. For this reason they are 
rightly regarded by chemists as much more permanent 
than beeswax, and Sir Arthur Church introduced them 
into the Gambler Parry medium in place of beeswax for 
this reason. Theoretically they would be better than 
beeswax for the mixture that I proposed for the backing 
of pictures, and they have proved excellent for the 
preservation of frescoes, being melted into the surface 
of the plaster. 

My reason for preferring beeswax as a backing for 
pictures is that the mixture with resin is sticky and 
firmly binds the two layers of canvas together. Paraffin 
wax does not form so sticky a layer. 

Gum-Arabic. Gum-arabic, which is obtained from 
the Acacia arabica, is used to prepare the medium for 
water-colour painting. It is an example of what is 
known as a vegetable gum, these gums differing from 
resins in being soluble in water. It was used as a medium 
by the Egyptians and forms a very durable film for the 



attaching of the pigments to paper, but has no protective 
value, and for that reason we have to select a more 
limited palette of pigments for the water-colour painter. 

There is another reason for limiting the palette, and 
that is that in water-colour painting the pigments are 
washed out in very thin layers exposing large surface 
without any depth to the action of air and light, so that 
a pigment which might prove quite suitable for painting 
in oil will not resist the severe test of being used in a 
water colour. It is for this reason that tests as to the 
permanency of a pigment are usually made in water- 
colour washes, so as to try them under the severest 

In the gi'inding of water colours, in order to keep them 
moist and to counteract the brittleness of the gum, it 
has long been the custom to add a Uttle honey, the use 
of honey for such purposes being mentioned by Cennino 
Cennini and in other early recipes. Honey consists of 
a mixture of two sugars, known as dextrose and laevulose, 
with a httle sucrose. The useful properties depend on 
the properties of laevulose. This is non-crystalline and 
gives the required flexibility. The laevuloses can be 
separated from honey by taking crystallized and semi- 
solid honey, mixing it with four times its volume of 
proof-spirits, filtering, and evaporating the solution. 

The objection to the presence of honey is that all 
sugars are slightly hygroscopic and attract moisture, and 
the presence of moisture is of course dangerous for the 
pigments, but, judging by the long tradition behind it, 
the presence of honey does no harm. 

The modern practice of the artists' colourman is to 
add glycerine in order to keep the water colour moist. 

Formerly artists were content to have their water 
colour prepared in cake form. Then they required them 
moist in pans ; now they demand them put up in tubes, 



each of these changes meaning an increase in the amount 
of the hygroscopic constituents, thus increasing the 
possibility of the fading of the picture. 

Size. In Chapter XIV I describe the methods given 
by Cennino Cennini for the preparation of size. Both 
the materials and modern methods of manufacture are 
very much the same to-day as in his time, the final 
product being a mixture in different proportions of two 
substances — gelatin and chondrin. 

I have already advised (Chapter V) that in order to 
obtain a pure and reliable article, the gelatin which is pre- 
pared for making solutions for the breeding of bacteria 
is the best. In order to prevent the size from putrifying 
a lump of camphor or a few drops of eugenol from oil of 
cloves is sufficient. 

A painting done wdth size, after it is dry, should be 
sprayed or painted over ^^^th a weak solution of formahn. 
This will make the size insoluble in water, although still 
capable of absorbing water and swelUng slightly. Such 
a surface of size treated with formalin is very permanent, 
and there seems to be no reason why size should not be 
used instead of egg in painting the foundations for an 
oil picture. As I have stated elsewhere, the oil film 
dries well on the size without cracking or wrinkling, and 
the two are firmly attached to each other. 

The follo^^'ing is a modern recipe for the preparation 
of parchment size from the Transactions of the Tempera 
Society, 1901-1907, by L. Agnes Talbot, taken from an 
account of how to gild on gesso to which the student 
who mshes information on this subject is referred. 

To MAKE Parchaient Size. The parchment shavings 
are cut into pieces and put to soak in water throughout 
the night. In the morning measure them in a vessel 
not pressed do^^Tl. Boil them in water (a double sauce- 
pan is best) until the smaller pieces begin to dissolve 




and break. Then strain through muslin. One measure 
(measured after soaking, before boiling) should make 
one and a half measures of size. If when you pour it off 
there is more than this, boil till it is reduced to the 
right amount, if less add water. This, I am aware, is 
not an exact or satisfactory method of measuring, and 
I should be very glad to know of a better. If one 
weighed the parchment dry in deUcate scales one could 
come nearer to accuracy ; but even then the continual 
warming of the size, when in use, increases the strength, 
and one has to add water, so that it becomes a rule-of- 
thumb business after all. This strength of the size is 
rather important, because on it depends the hardness 
of the gesso ; a too soft gesso does not take burnish and 
is tiresome to manipulate, while on a too hard one the 
gold loses quality and has a glittery appearance. 

Casein. Casein is prepared from milk, and is one of 
the constituents of cheese, which consists of casein and 
the butter fats that milk contains. It is a nitrogenous 
body corresponding closely to size and albumen, and 
has been introduced in the isolated state as an article 
of commerce in comparatively recent times. It is 
insoluble in water, but can be dissolved by the addition 
of a little ammonia, potash, or lime, and then ultimately 
hardens and becomes insoluble again. It is very largely 
used to-day for the manufacture of water paints — 
casein, a little lime, and the pigments being ground up 
together, and it has been used by artists both as a paint- 
ing medium and for the preparation of grounds. The 
safest solvent is ammonia, as this evaporates, lea\'ing 
the casein behind ; alkalies, such as potash, remaining 
in the film, which may ultimately prove injurious. 

If used for the preparation of grounds or of a painting 
medium, it dries very hard and brittle, and, therefore, 
the introduction of glycerin has been advised. The 



introduction of glycerine is objectionable, because it is 

It must be remembered that our experience of casein 
is short ; but, on the other hand, as far as our experience 
goes it has proved reliable for the preparation of grounds. 
On the whole, therefore, I prefer to advise preparing 
grounds either with size or oil, thus sticking to known 
tradition. It is more doubtful as a painting medium. 
Very complex recipes have been made up containing 
caustic potash, castor oil, and other ingredients. The 
presence of the caustic potash will, to some extent, 
saponify the oils introduced, and it is impossible for 
anyone to predict what the future behaviour of such 
complex mediums will be. 

It has been argued that because there are mediaeval 
recipes for cementing together the different portions 
of a panel vdth cheese and lime, casein must be a reliable 
substance ; but the cheese contains the butter fats, 
which ^^dll be saponified by the lime as well as the casein, 
and such a combination may have proved very durable 
as a cement, but it is never recommended either for the 
making of gesso ground or as a painting medium, and 
the presence of the butter fats may have a protective 
value. There are very impure caseins on the market, 
and only the best quahty should be used. It is nearly 
white, has no smell, is not acid to litmus paper, and 
gives a nearly colourless transparent solution with water 
and ammonia. 

There are other gums and materials such as starch 
which can be used as mediums for painting, but they are 
not commonly used. 




The medium used universally in Italy during the four- 
teenth and, at any rate, up to the middle and to about 
the close of the fifteenth century, was yolk of egg. 

It is difficult to say how old the egg medium is. We 
know that the Greek painters of classical times used two 
media, and that one was wax ; but we have no reUable 
record of what the other medium was. Pliny, while 
describing the pigments used and the technique of the 
wax medium, states that two media were used, but 
omits to mention what the second medium was. It 
may have been gum-arabic, used largely by the 
Egyptians, glue, or egg, or a drying oil. Pliny mentions 
the use of egg, but only for specific purposes. The 
Lucca manuscript of the eighth century mentions size 
and wax as media, but makes no mention of egg. As 
we have already stated, in the manuscripts of the 
eleventh and twelfth centuries, we have descriptions of 
oil painting, but no mention of painting with yolk of 
egg. The statement of the early and universal use of 
egg yolk as a medium seems to have been copied from 
Vasari, and it is, at any rate, open to question whether 
it was universally used in earlier centuries than the 
thirteenth, though it is highly probable that so excellent 
a natural medium was devised in early times. 

We have sufficient evidence from the excellent con- 
dition of many pictures of the fourteenth and fifteenth 

M 177 


century, painted in this medium, to satisfy us as to its 
lasting qualities, and its value as allowing pigments to 
retain their full brilhancy of tone ; and we are fortunate 
in possessing in the treatise by Cennino Cennini, sup- 
posed to be "WTitten about 1432, a most complete and 
detailed account of this process, so that the artist 
painting in yolk of egg to-day has a full guide to all the 
technical processes involved. 

Before proceeding to give a short resume of this 
method of paintmg, as described by Cennino Cennini, 
I think it advisable to say something of the special 
properties of yolk of egg as a painting medium. The 
white of the egg consists of an emulsion of egg albumen 
and water, which was used for attaching gold leaf and 
in miniature painting. The yolk is an emulsion con- 
taining, among other things, water, albumens, an oil, 
and a substance known as lecithin. 

The oil consists of the glycerides of oleic, palmitic, and 
stearic acids. It is therefore an example of what we 
should call a non-drying oil, like olive oil ; but as an 
actual fact these fatty oils will dry in course of time, 
and we cannot make a sharp distinction between a 
drying oil like linseed oil and a non-drying oil like olive 

The egg albumen if raised to 70°-75° C. becomes 
insoluble in water, and if exposed in thin layers at 
ordinary temperatures, ultimately becomes insoluble. 
We have here an example, therefore, of what we call 
an emulsion containing an oil, containing water, and 
containmg an emulsifying agent, namely, the albumen, 
in which in course of time the albumen will become 
insoluble, and the oil will " dry," so that we obtain a 
protective lajer, which is very insoluble in any agent 
and very durable. 

In addition to these substances the yolk of egg 



contains a very remarkable substance known as lecithin, 
which is a fatty body which swells up when moistened 
with water, and therefore is an ideal agent for emulsi- 
fying oil and water together. This substance, lecithin, 
is found widely distributed through both the animal 
and vegetable kingdom, and seems to be nature's 
emulsifying agent. Probably if isolated and prepared 
for artists' use, it would prove very valuable. We can 
therefore regard yolk of egg as the most perfect of oil- 
water emulsions. 

To return to Cennino Cennini, he begins with the 
preparation of the panel, and follows the whole process 
of painting a picture right through. We have already 
dealt fully with the preparation of the ground in 
Chapter III. 

We therefore now come to a discussion of the actual 
mediums to be used in painting on this panel. Cennino 
Cennini describes three — size prepared from parchment, 
which he recommends for laying on real ultramarine ; 
linseed oil refined and bleached in the sun, which he 
recommends for painting on gold with viridian and lake, 
and also for painting the pile of velvet ; and, thirdly, 
yolk of egg. He advises the use of the white well 
beaten and mixed with water for the laying on of gold, 
and either the use of the yolk and the white beaten up 
with the young shoots of the fig tree, or, as a better 
medium, the yolk alone. 

The reference to the fig tree and, in another recipe, 
to the addition of fig juice, where the egg is to be used 
to prepare the surface of a wall for tempera painting, 
has given rise to much speculation. It has been sug- 
gested that as the fig tree belongs to the same family 
as the rubber tree, the juice probably contains some 
caouchouc. Now that the rubber latex is obtainable in 
this country, experiments with mixtures of the latex 



and egg would be of interest. Eibner thinks that the 
fig tree juice was simply added as an antiseptic in place 
of wine or vinegar, which are both mentioned in old 
recipes, and it is also possible that in the case of a 
mixture of the yolk and the white, it made the medium 
flow better under the brush. 

It is sufficient for our purpose that Cennino Cennini 
recommends the pure yolk of the egg as the better 
medium ; and he also recommends town eggs, as he 
says the yolk is lighter in colour than country eggs. 
The yolk mixed with water is ground on a muUer with 
pigment. He tells us to add about equal quantities 
of yolk to pigment ; but the amount varies with every 
pigment, and must be learnt by experience. His 
pigments are first ground with water and kept under 
water ready for use. He does not tell us how much 
water to add with the egg, but the pigment must have 
been very wet. 

It is evident from his account that painters in Italy 
of his time did not necessarily confine themselves to 
egg, though it was their principal medium. 

Finally, Cennino Cennini advises us to varnish the 
picture by rubbing over hot varnish with the hand, but 
this is not to be done till the picture is at least a year 
old. In spite of Cennino Cennini's directions, I do not 
believe that this was always done. It is obvious that 
in many cases the varnish is no longer there, and it has 
been suggested that it may have crumbled away. I 
have examined carefully tempera pictures which are 
in excellent preservation, and have never been touched 
by the restorer or covered by a modern varnish, and can 
see no indication of their having ever been varnished 
unless possibly with white of egg, which Cennino Cennini 
recommends as a temporary varnish. 

If a picture is painted correctly in yolk of egg, it can 



afterwards be polished with a soft rag, taking an egg- 
shell gloss, and this is exactly the appearance that many 
of these old pictures have. Modern tempera pictures 
more than twenty years old have the same glossy 
surface. The most perfect pictures in Italy come froni 
the nunneries, where they have been taken care of in 
a clean and perfect atmosphere, and seem to be just as 
they left the studio, without any indication of change. 

Successful painting in tempera requires considerable 
practice and experience ; the pigment has to be laid on 
very thin ; each thin layer must have time to dry ; 
and the amount of yolk of egg present must be neither 
too much nor too little. 

Cennino Cennini says nothing about preventing the 
yolk from putrifying. Other old recipes mention the 
addition of wine or vinegar. The explanation is simple. 
The egg yolk is slightly alkaline, and putrifactive 
bacteria favour an alkaline medium. Either vinegar or 
the acids of the wine will therefore, by making the egg 
yolk acid, prevent this ; and the alcohol of the wine 
will also act as a preservative. 

Sir Arthur Church recommends a saturated solution 
of eugenol, prepared from oil of cloves in 5 per cent 
acetic acid. This is added drop by drop with constant 
agitation to the yolk until turmeric paper, which is 
turned brownish red by the alkaline egg, just turns 
yellow again. A piece of camphor should also be 

An acid solution will attack ultramarine blue. This 
blue should therefore not be used if the yolk has been 
acidified. Possibly the reason why Cennino Cennini 
recommends size for real ultramarine, which is also 
destroyed by acid, is due to the fact that he did add 
vinegar, but has omitted to say so. 

I shall now give Mr. Duncan's account of how to paint 



in tempera ; I have already given his method of preparing 
his panel. The pigments are all ground in water, and 
placed in little covered jars. The water is allowed to 
settle and the excess of water poured off. The yolk is 
separated from the white. 

Here I insert Mr. Tudor Hart's method of preparing 
the yolk of egg. Having separated it from the white 
he removes excess of white of egg by rolling the yolk 
gently from the palm of one hand to the palm of the 
other, and wiping the free palm between each change. 
Pinching the enclosing skin of the yolk between the 
thumb and finger, and lifting it gently from the palm 
of the hand, holding it suspended over the grinding 
slab, with a thin pointed blade he punctures the yolk 
from underneath by a quick stab and let the yolk run 
out. The rest of the yolk can be expelled by placing 
the deflated skin on the slab and pressing it with the 
finger from the point at which it is held to that of the 

To continue with Mr. Duncan's account, add to each 
egg two egg-spoonfuls of water ; strain the whole 
through muslin ; and to every two eggs add one egg- 
spoonful of a 3 per cent solution of acetic acid. This 
will keep the egg fresh for a week, even in warm weather. 
Equal volumes of egg and pasty pigment are ground 
together on the muller. Some pigments, like terre verte 
and yellow ochre, require more egg. If a little of the 
pigment and its egg medium is painted out on glass 
and left to dry, next day it should be possible to remove 
it with a spatula as a leathery film. If it is powdery, 
enough egg has not been added. If the panel has been 
properly prepared, and the pigments thoroughly ground, 
each layer of pigment, when dry, can be polished with 
a piece of soft linen rag, giving translucent colour with 
an egg-shell gloss. If the pigment remains dry and 



dead and opaque, the gesso has not been properly pre- 
pared. The finished picture is polished with a rag and 
does not require varnish, but after thorough drying can 
be varnished with mastic. Overlapping can be done 
so as to avoid the necessity of hatching or stippling. 
Thin pigment and the brush squeezed out between each 
operation, as directed by Cennino Cennini, are essential. 
Preliminary shadows should be done with verdaccio. 
If it is desired to paint freely and easily with ready 
blending of tints, an emulsion of egg with oil or copal or 
mastic varnish must be used. This medium works 
excellently, but yellows slightly with age, and is apt 
to be easily separated from the gesso below. This is 
probably due to the use of oil copal. Gesso grounds 
on canvas crack and are not satisfactory. Casein 
grounds on canvas also crack, and are very hard and 
brittle. A tempera painting kept in the dark for fifteen 
years is as bright as the day it was painted. 

We now come to the method adopted by Cennino 
Cennini to build up a picture which is of supreme 
importance, as throwing light on the early technique 
adopted in oil painting. He describes a pigment of a 
neutral tint, which he calls verdaccio, and which is a 
mixture of white, black, yellow ochre, and red ochre, 
giving a useful neutral tint, which he uses both for 
fresco and for tempera — the white in the one case being 
his special preparation of whiting, and in the other case 
white lead. With this pigment, as Lady Herringham 
points out, both in fresco and in tempera, he lays in 
his whole scheme in monochrome. On this thin glazings 
of bright pigment mixed with white are laid. 

In the chapter on Oil Painting the early oil technique 
is explained, and its significance when we take full 
consideration of the optical properties of the oil medium. 
Such a technique was not so essential for either fresco 



or tempera, but seems to have been common at this 
time to all schools of painting. 

Cennino Cennini advises, in the case of tempera 
painting, washing a thin uniform tint of terre verte as 
an under-painting on the flesh parts of the picture, thus 
lowering the tone of the flesh by diminishing the reflec- 
tions from the white gesso through the thin painting 
above. It seems to have been assumed that this was 
the universal practice of tempera painters. Judging 
by the high luminosity of their flesh, I doubt if this is 
true. It is certainly not true of any early oil pictures 
I have examined by taking a section right through the 
flesh. I have referred in the chapter on Oil Painting 
to the unfinished Michael Angelo in the National 
Gallery where this has been done, and to the consequent 
low tone of the flesh painting. 

I have not discussed in this chapter the laying of gold 
leaf on a gesso panel. The readers can refer to Lady 
Herringham's translation on Cennino Cennini and the 
article on Gilding Gesso by L. Agnes Talbot in the 
Transactions of the Tempera Society, 1901-1907. Nor 
is it within my province to deal vnih the problem of 
painting in tempera and yet getting the fusion of tone 
without hatching, so successfully obtained by Jacopo 

I suspect the use of oil paint, to supplement the 
work in egg, to have been done much more frequently 
by the Italian tempera painters than has been supposed. 




The re\dval of painting in yolk of egg has led to an 
increased interest on the part of painters in emulsion 
mediums. I have already explained that yolk of egg is 
the most perfect of emulsions. An emulsion medium 
consists of an oil or varnish, water, and an emulsifying 
agent. Among the emulsifying agents are size, albumen, 
gum-arabic, gum-tragacanth, casein, and soaps. 

The artists' colourmen have put on the market tem- 
pera colours in which, in many cases, artificial emulsions 
are used. Some artists like to prepare their ovm emul- 
sions, and oil-size emulsions are largely used by the 
house-painter, under the names of water-paint and 
washable distemper. 

The advantage of an emulsion is that it enables the 
artist to paint freely \\'it}i water as a diluent, and at the 
same time to have his pigments bound with a more or 
less waterproof medium. The pigment dries flat or dead, 
and owing to the low refractive index of the medium, 
the brilliancy of the pigment is assured. The emulsion 
seems to have the property, while drying, of preventing 
the oil floating to the surface. 

In order to make a reliable emulsion, the proportions 
between oil, pigment emulsifying medium, and water 
have to be carefully adjusted. The medium must bind 
well, forming a waterproof coating, and the pigment 
should not crack when laid on thickly. The paint layer 
must remain firmly attached to the priming. 



It is in the matter of this last requirement that we 
have not, in my opinion, sufficient information. Vasari 
tells us that experiments with emulsions were abandoned 
before his time because the paint tended to become de- 
tached from the ground ; and I have noticed the same 
thing in the case of copal varnish, egg emulsion ; and 
the house-painter sometimes has the same experience 
with water-paint. I am not satisfied, therefore, that we 
have sufficient evidence of the reliability of emulsions 
outside the natural yolk of egg emulsion, and probably 
also a natural milk emulsion. 

As I have already pointed out, the yolk of egg emul- 
sion is unique, as it contains lecithin, a fatty substance 
which emulsifies with water. For that reason, the best 
substance available for making emulsions is the 3'olk of 
egg itself. Blended with a little oil or varnish, it be- 
comes possible to paint freely with it, thus producing a 
medium which is much more facile to the needs of the 
painter than egg alone. But, as I have stated above, it 
is possible that it may peel off, especially if laid on in 
thick layers. 

It is evident that any mixture of, let us say, oil, water, 
and size will not form a satisfactory emulsion. They 
must be proportioned to each other in such a way as to 
obtain an easily worked medium which can be mixed 
with water. The first principle to be laid down is that 
the oil content must be sufficient alone to bind the 
pigment. The amount of oil required to bind the pig- 
ment is much less than the amount required to grind 
the pigment. If equal weights of pigment are ground 
with different quantities of oil, thinned with turpentine 
to make grinding possible, we can easily determine the 
minimum amount just necessary to bind. 

As an example of the adjusting of these four propor- 
tions we may take an emulsion prepared for water-paint. 



The pigment used was lithophone. By experiments ynth. 
boiled oil and lithophone in the way prescribed it was 
found that ten parts by weight of hthophone was just 
bound by one part of oil. This gave the minimum of 
oil it would be safe to use, and starting on this as the 
minimum quantity, a satisfactory emulsion was ulti- 
mately obtained ^\ith fifty parts of lithophone, seven 
parts of oil, two parts of size, and twenty of water. 
This emulsion, while making a satisfactory water-paint, 
would crack if laid on thickly, o%\ing to the high water 
content. If the water is kept do^^-n, so as to diminish 
the danger of cracking, the amount of pigment requires 
also to be reduced. 

When making an emulsion ^^'ith yolk of egg and oil 
the addition of a httle oil of turpentine, or oil of spike, 
assists the amalgamation. 

In the chapter on Oil Painting I have discussed the 
possibility of the use of Dutch Stand oil by the early 
painters. An emulsion of one part of yolk of egg by 
volume to one part of Dutch Stand oil, and a little 
turpentine, gives an excellent medium which, when 
ground with white lead, gives a crisp and easily worked 
paint which dries %\'ithout flo's^'ing. Such a medium 
cannot be diluted mth water. 

In order to make an emulsion vdth Stand oil which 
can be diluted %\'ith water, two parts by volume of egg 
mixed with three parts by volume of water to one part 
by volume of Stand oil are required. Neither of these 
emulsions crack on drying when painted on thick. 

A properly balanced egg oil emulsion can be mixed 
with oil, water, or turpentine. Balsams amalgamate 
excellently -snth yolk of egg. 

Two volumes of egg yolk, one volume of linseed oil, 
and three volumes of water form an excellent emulsion 
that can be mixed with water. 



Among the materials which can be used as emulsiMng 
agents, as we have already stated, are the following : 
Size, gum-arabic, gum-tragacanth, casein, albumen, yolk 
of egg, and soaps. Of these, soaps are to be absolutely 
condemned, and casein is of very doubtful value. In 
order to dissolve it, an alkali — soda, potash, ammonia, 
or hme — must be added ; and this alkali vdW form soaps 
wdth the oil, thus introducing a dangerous element. 
Moreover, we have no experience of casein over long 
periods of time, and it is the least stable of the nitro- 
genous group to which it belongs. 

There are many tempera mediums on the market, 
of which some, if not all, are artificial emulsions. There 
are so many ways of making up emulsions, and so many 
of them are unsound, that artists' colourmen should be 
required to state the composition of the medium on the 

If an artist proposes to use one of these tempera 
mediums, it should stand the following tests : In the 
first place, a little of it moistened with water should be 
tested with litmus paper. If it is strongly alkahne, it 
probably contains a soap, though it is just possible that 
the alkalinity may be due to vrhite of egg. It should be 
rejected unless the artists' colourman will guarantee the 
absence of soap-emulsion. 

If it passes this test, a little should be rubbed out with 
water on a sheet of glass, and a lump squeezed out of 
the tube on to the glai^ as well, and the whole put aside 
for a month. At the end of that time it should have 
formed a tough film on the glass, and the lump should 
have dried -uithout cracking. The glass sheet should 
now be plunged into hot water and left to stew in hot, 
not boihng, water for some hours. At the end of that 
time neither the film nor the lump should have dis- 
integrated or become separated from the glass. 



In discussing the Van Eyck green in my book on 
Pigments and Mediums of the Old Masters, I pointed out 
that the only green I knew of, available in the early 
fifteenth century, to produce the effects to be seen in 
these pictures, was made by dissolving verdigris in a 
pine balsam, and the resulting product would be too 
sticky to paint with, unless diluted with oil of turpentine, 
and that, as far as we could tell, this medium was not 
used by painters at that time, and that the only way 
to make a workable medium would be to emulsify the 
balsam with egg and a little oil. 

It seems to have been assumed that I intended an 
emulsion that would mix with water. There is no need 
to make such an emulsion. If a little yolk of egg is 
mixed with a sticky oil like Stand oil or with varnish, 
a medium is obtained which will not mix with water, 
but will paint out nicely, leaving firm, crisp strokes 
which do not flow. That in certain cases the early 
fifteenth -century oil painters used such methods seems 
to me highly probable, but that does not mean that they 
prepared media which were miscible with water. For 
instance, one part of egg yolk to one part of Stand oil 
by volume produces a medium which gives a firm, crisp 
consistency to white lead and dries without flowing. It 
is essentially an oil medium, only containing about 7 per 
cent of albumen. If verdigris is dissolved in Venice 
turpentine, and one part of Venice turpentine by volume 
be mixed with one part of linseed oil, and one part of 
yolk of egg, a satisfactory medium is obtained which, 
while opaque when first painted out, becomes trans- 
parent on drying, giving a most beautiful green. 

It has been stated more than once, even by painters, 
that the Van Eycks used a varnish medium. Varnish 
mediums are sticky and flowing, and it is impossible 
that the early Flemish pictures could have been done in 



such a medium. In fact, the strokes of the brush are, 
in many cases, possibly too precise even for oil alone. 
In pictures of a later date little raised lumps of paint 
are often seen in the painting of jewellery and such-like, 
which are so sharp and definite as to suggest some 
amalgamation of the oil. The grinding of the pigment 
in oil or varnish with the yolk of egg gives the required 

This suggestion of mine is quite different from the 
water-emulsion theory of the painting of these pictures 
at present in vogue, though a water emulsion may have 
been used in place of egg or size for the under-painting. 

Finally, the safest emulsion for artists to use is yolk 
of egg, and if more freedom of handling is required, it 
should be obtained by the addition of a minimum of oil. 

The reader will find many interesting experiments on 
emulsions in Professor Berger's Bcitrage zur Entwickel- 
ungs Geschichte der Maltechnik. There is also an 
interesting paper by Mr. Tudor Hart in the Transactions 
of the Tempera Society, 1901-1907. 




Of all the methods of painting, Buon-Fresco exercised 
the most powerful fascination over the Italian artists of 
the Renaissance. It was regarded as the supreme test 
of skill, requiring the most perfect craftsmanship, while 
utilizing the simplest materials. 

In Buon-Fresco the pigments mixed mth water are 
painted on the wet plaster of sand and lime, and are 
finally cemented into position by the conversion of the 
lime, by combination with the carbonic acid gas in the 
air, into calcium carbonate. 

The most interesting and valuable experiments made 
to revive the process are to be found in the Houses 
of Parliament. IVIrs. Merrifield made an exhaustive 
enquiry into such information as was available of 
the ancient methods, and contemporary artists were 
employed to carry out a big scheme of decoration. 

From the point of view of permanency the results 
were unfortunate, repeated restoration and fixing by 
means of solutions of paraffin, wax, and varnish ha\'ing 
been necessary. The rapid decay of these frescoes has 
confirmed the conclusion of chemists that we should 
avoid Buon-Fresco for decorative purposes in our 
modern cities, with air laden with soot and with the 
acids produced by the burning of the sulphur in coal. 
Against the views of the chemist modern followers o 
the process claim — at any rate, in the country districts — ■ 



that modern frescoes have sho\Mi considerable per- 

^^'hile I am in agreement -vvith my fellow-chemists in 
condemning its use in our modern cities in this country, 
there is no reason why it should not be used elsewhere. 
The cities of Italy are practically as free from coal smoke 
as they were in the days of Raphael ; and America, with 
its dry climate and absence of smoke in many of its 
cities, is peculiarly suitable for fresco painting. It must 
also be remembered that a revolution in the use of coal 
is rapidly approaching, o^nng to the universal applica- 
tion of electric current for light and power, while the 
necessary economy of coal -wdll compel us to abandon 
our present wasteful methods, and we may hope in the 
near future to have smokeless cities. 

There may therefore be a great future before fresco 
painting. An examination of the frescoes in Italy makes 
it difficult to come to a conclusion as to the durability 
of the method under proper conditions. In many cases 
they are in excellent condition, in others they have been 
restored, and in Others they have very largely dis- 
appeared. I suspect that it depends on whether the 
walls themselves are dry or damp. The damp course of 
the modern architect was unknown in those days, and 
the soaking up of water with the dissohing and re- 
crystallization of the lime compounds present may be 
the cause of decay. There may also have been periods 
of exposure ouing to neglect of the buildings. On the 
whole, the evidence is in favour of the \'iew that 
under proper conditions, in a cUmate like Italy, and 
in towns free from coal smoke and sulphur acids in the 
air, they are a remarkably permanent form of decora- 

It has, however, been assumed, because of the bad 
state of decay of " The Last Supper," by Leonardo da 


lU'oX FRESCO. By Benozzo GozzoLi. 
A portion of the fresco iiainting by Benozzo Gozzoli in the Chapel of the Riccardi, Florence. 

BrON FRESCO. By Benozzo Gozzoli. 

Anothtr example of frjsco pointing by Benozzo Gozzoli in the Chapel of the Riccardi Palace 



Vinci, that oil forms a fugitive form of wall painting. 
If tradition is correct, there are examples of oil painting 
in the Vatican frescoes which are in perfect preserva- 

A matter of the first importance is the careful prepara- 
tion of the plaster surface on which the painting of the 
fresco is to be done. The Romans took elaborate pre- 
cautions in the preparations of their plaster surfaces, 
which were some 5 in. thick, and which have been very 
fully described by Vitru^dus, an architect who lived in 
the time of Augustus. The first preparation was a 
coarse concrete of broken brick and lime followed by 
lime and sand, and finally finished with two or three 
coats of lime and marble dust. Care was taken that the 
slaked lime was thoroughly matured, and the mortar 
was in the first place stamped in the tub, and then 
worked and beaten on the wall. A stiff paste of lime 
belongs to that variety of semi-solid bodies which, while 
it sets up stiff, can be broken down into a sludge, which 
again sets up stiff — plasterers call this process " knock- 
ing up " — so that by proper breaking down it can be 
got into a usable condition without the addition of too 
much water. Each coat was laid on before the last was 
thoroughly dry, and the final coat, which, if a uniformly 
covered surface was required, was mixed with pigment, 
was worked with a trowel until a highly polished sur- 
face was obtained. On this surface the painting was 

There has been much dispute as to whether this final 
painting was done with the pigment mixed with water 
or water and a little lime, or whether some medium like 
wax was used. The evidence, on the whole, is in favour 
of Buon-Fresco, the pigments being merely mixed with 
water, or possibly with a little slaked lime. 

One of the objections to the view that they were 

N 193 


painted in Buon-Fresco is the absence of joins in the 
Pompeiian frescoes ; but it must be remembered that 
the great thickness of the plaster and the poUshing of 
the surface would enable it to be kept moist for days. 
I have found that it is possible by proper working with 
the edge of the trowel day after day as the plaster 
gradually dries, with a mixture of lime and marble dust, 
to get up a surface like polished marble. No attempt 
seems to have been made in modern times to revive this 
classical method of fresco painting. I have also found 
that pigments laid in water on a surface of lime and 
marble dust, and then lightly pressed home with a 
trowel, became firmly attached and could not be re- 
moved by water ; and that if the Roman practice was 
adopted of a thick plaster damp throughout and 
" closed " by trowelling, then the fresco could be laid 
on some forty-eight hours after the plaster had been 

In certain cases there is no doubt that beeswax was 
driven in by heat and polished as a final protective 
varnish, but there is no evidence of its universal use. 
It ought to be remembered that the Greeks and Romans 
admired polished surfaces. 

While there is still much that is obscure in their 
methods, the care with which they prepared the plaster 
is well worthy of imitation. Marble dust and lime make 
an excellent finishing coat, and the gradual building up 
from the coarse broken brick to the sand, and from 
the sand to the still finer marble dust, is thoroughly 

There is little of practical value to be learnt from the 
accounts we possess of the methods used in the Middle 
Ages, and the next information of value is to be obtained 
from the treatise of Cennino Cennini. Cennino Cennini 
describes three methods — painting on walls, iji secco with 



egg or \\dth oil, and painting by the wet method or Buori' 
Fresco. We shall in the meantime confine ourselves to 
a description of Buon- Fresco. 

It is evident that by his time a complete technique 
had developed, differing from the methods described by 
Theophilus and in the Hermeneia. The wall ha\ing been 
plastered, a drawing of the proposed fresco was made on 
the dry plaster of the wall, and then each day a freshly 
prepared plaster of lime and sand was spread over the 
portion which would occupy a day's painting, and on 
this wet surface the pigments were laid tempered with 
water. At the end of the day any portion of the plastered 
surface not utilized was cut away. The description of 
fresco painting given by Vasari does not differ in essen- 
tials, the preparation of a cartoon and pouncing on 
plaster ha\dng been substituted for the older method of 
drawing on the wall direct. Cennino Cennini describes 
very fully the preparation of the various tints and their 
utilization, the laying out of the wall surface, and 
so on. These matters it is not my province to discuss 

The success of Buon-Fresco depends principally upon 
two things — the careful preparation of the slaked lime 
and the thorough working and mixing of the mortar. 
Quicklime is prepared by the heating with fuel in a kiln 
of carbonate of lime, of which many varieties occur in 
nature, such as chalk, limestone, and marble. The car- 
bonic acid gas is driven off, leaving quicklime or calcium 
oxide behind. This compound is capable of combining 
with water to form slaked lime or calcium hydrate. It 
is obvious that, if these processes are considered, it is 
essential that the quicklime should be thoroughly slaked. 
Any particles not thoroughly slaked will slowly absorb 
water and, as they slake, expand and crack the plaster. 
Even quicklime prepared from marble contams some 



portions that are over-burnt and others which contain 
impurities, which will therefore slake very slowly. 
. Quicklime should therefore be obtained from as pure 
a source as possible. Marble is certainly the best ; but 
Buxton lime is very pure. It should be slaked with 
water, and with occasional stirring and turning over, 
and kept for at least three months before use. Some 
people prefer to keep it for years, but if kept so long, 
care must be taken to exclude as much as possible 
the air which contains carbonic acid gas, and which 
therefore by carbonating the lime will destroy its 

There is considerable difference of opinion as to the 
right proportion between sand and lime. Pacheco and 
Palomino recommend half and half. Cennino Cennini 
recommends one-third lime to two-thirds sand. "Wlien 
thoroughly worked and amalgamated with sand, the 
trowel should come away clean from the mixture. Lime 
already slaked is now obtainable as a commercial pro- 
duct, and this should be a good starting-point, being 
mixed with water and allowed to mature instead of 
starting with quicklime. Clean sand should be used to 
mix with the lime, free from impurities and approaching 
as nearly as possible the chemically pure silica or quartz. 

Ground quartz is now obtainable as a commercial 
product, and the introduction of a certain amount of 
this as a filler is beneficial. It is known that a much 
stronger and more durable concrete can be made by 
mixing cement with a graded aggregate, starting from 
coarse particles and ending with very fine ones. The 
sand should also be graded for the plaster on which the 
fresco is to be painted by sifting, in the first instance, 
into a series of different degrees of fineness. The follow- 
ing example of the grading of aggregates for the making 
of asphalt pavements should prove useful. The sand for 



the asphalt pavements in New York is graded as 
follows : — 

Passing 100 mesh to the inch sieve = 13 per cent 
80 „ „ „ = 13 


= 23-5 

= 110 

= 80 

= 50 

= 30 

If the reader refers back to the description of the 
Roman method of preparing a plaster wall, he will see 
that the Romans graded from the back to the front of 
the plaster, using broken bricks for the first layer, and 
finishing with marble dust. Pieces of flint should be 
wedged into the wall to attach the first layer of plaster, 
and asbestos fibre should be introduced throughout. 

Great care should be taken with the damp course, and 
where the fresco is to be on an outside wall a brick wall 
should be built inside the outside wall with an air space 
ventilated from the inside of the building. These pre- 
cautions should be taken whatever method of fresco 
painting be adopted. 

Pigments Suitable for Fresco 

Cennino Cennini gives a very limited list of pigments. 
A white, bianco sangiovanni, of which I shall say more 
presently ; black, the yellow, red and brown ochres, and 
terre verte, which is also a natural earth owing its colour 
to compounds of iron ; and ultramarine from lapis lazuli. 
The ultramarine, he says, must be laid on in secco with 
an egg medium. The Italian painters up to 1600 had 
only two blues, real ultramarine and Azzuro della Magna 
— ^that is, azurite, the blue carbonate of copper obtained 



from mines in Hungary. They did not regard it as safe 
to use real ultramarine on the wet lime ; and azurite will 
not stand the action of wet lime. 

Towards the close of the sixteenth century smalt was 
introduced. The Melozzo da Forli fresco in the Picture 
Gallery of the Vatican is painted with smalt, but I 
suspect a restorer's hand replacing an older blue. 

I suspect therefore that even in the time of Raphael 
the blues were laid on in secco. That azurite was used 
in large quantities is proved by the ceiling of the Fran- 
ciscan Church at Assisi. The blue has turned in many 
cases to a magnificent green, evidence of azurite. 

Terre verte, raw sienna, and yellow ochre, according 
to Professor Church, are apt to cause disintegration of 
the surface. I noticed terre verte doing so in the frescoes 
in the library at Siena. There is some doubt as to the 
right yellow ochre to use. Pliny suggests that a special 
ochre, which he calls " marmorosum," should be used. 
For this reason. Ward suggests replacing ochre with 
cadmium yellow, but, apart from the fact that it is quite 
the ■v\Tong colour key to go with these other earth colours, 
cadmium yellows are too much open to suspicion to be 
used for this purpose, and enquiry should be made into 
the properties of the earth pigments to be used. 

It is quite evident from the frescoes in Italy that 
artists had no difficulty in finding suitable earth pig- 
ments, including yellow ochre. Possibly the same 
sources of supply could be used. 

The modern artist is much better equipped. He has 
viridian, and green oxide of chromium and cobalt blue, 
cerulean blue, and smalt for blues. For the rest he 
should confine himself to the red, yellow, and brown 
earths. Potter's pink and potter's vernalis are safe. 

So far I have said nothing about the white to be used 
in fresco painting. The modern and, I believe, the 



sixteenth-century practice, was to use lime so that a 
binder and a white are being used together. Cennino 
Cennini recommends a white which he calls bianco 
sangiovanni, and which he prepared by moulding the 
lime into little bricks and leaving them exposed to the 
air for some months. This would result in the partial 
carbonating of the lime, the result being an intimate 
mixture of lime and chalk, which will be whiter than 
pure lime. 




As an example of modern practice I reprint here the 
greater part of the paper on " Fresco Painting," by- 
Mrs. Sargant Florence in the Transactions of the Tempera 
Society. Students should consult a further paper by 
Mrs. Florence in the Transactions of the Tempera Society 
for 1907, 1924. She has finally adopted a mixture of 
one part fat lime, one-half part pit or river sand, and 
one-half part marble dust. 

She uses the earth colours : vegetable black, bianco 
sangiovanni (refined lime putty), potters' pink, viridian, 
chromium oxides, and potters' " vernalis," also an 
oxide of chromium preparation, cobalt blue, artificial 
ultramarines, and perigord yellow (a yellow ochre). Of 
these the only one I regard as doubtful is ultramarine. 

" To define the nature of fresco painting I cannot 
do better than quote Vasari, who puts the matter into 
a nutshell. He says : ' The picture must be painted 
on the lime while it is wet. The work must not be left 
until all that is intended to be done that day is finished. 
Because if the painting be long in hand a certain thin 
crust forms on the lime, as well from the heat as from 
the cold, the wind and the frost, which tarnishes and 
spots all the picture. And therefore the wall which is 
painted upon must be continually wetted ; and the 
colours employed upon it must all be earths, and not 


EVE. Bj- Michael Axgelo. 
An example of fresco painting bv Michael Angelo taken from the Sistine Chapel in the 


minerals, and the white must be calcined travertine. 
This kind of painting also requires a firm and quick 
hand, but above all a good and sound judgment, because 
while the wall is soft the colours appear quite different 
from what they do when the wall is dry. It is therefore 
necessary for the artist, while painting in fresco, to use 
his judgment more than his skill, and to be guided by 
experience. . . .' 

" The subject naturally divides itself into four 
parts : — 

"1. Preparation of the wall — tools. 
" 2. ,, ,, cartoons. 

" 8. „ „ colours. 

" 4. Laying on of colours — tools. 

" Preparation of Wall Surfaces 

" In the preparation of the walls I have been obliged 
to base my practice altogether on the traditions handed 
down by old writers, as I have not had the good fortune 
to come across any masons who have had experience of 
that kind of work. These traditions apparently vary 
very considerably, but I gather that the disagreement 
lies mostly in the number of coats laid on rather than 
in the material. Wall plaster always seems to consist 
of slaked fat lime, with some gritty material which it 
binds together ; and a wall surface properly prepared 
for fresco, is composed of three different kinds of plaster 
which vary according to the kind of grit. There is (1) 
the Undercoat (Italian : Trullisatio) of lime and pounded 
brick (size of peas) or baked pottery, laid on coarsely 
on laths or bundles of rushes carefully laid in place and 
left rough to serve as a base for the upper coats ; (2) 
the Lime and Sand Coats (Arricciato) ; (3) the Lime and 
Marble {Intonaco). The last one receives the painting* 



"According to Vitruvius, one coat of trullisaiio was 
sufficient, whilst three of the arricciato or sand coats 
and the same number of intonaco or marble were neces- 
sary. Probably this verj^ elaborate preparation was for 
surfaces which we associate with Pompeiian style, half 
of whose beauty is due to the fine texture of the plaster ; 
and which was made to be decorated in flat spaces of 
colour, as he speaks of burnishing and polishing the 
intonaco, in order to enrich the quality of the pigments. 
The instructions of Italian artists such as Alberti and 
Vasari, and of Guevara among the Spanish, are based 
mostly on Vitruvius, who, in Book ^^I, Chapter 3 
(Gwilt's translation), gives a very detailed and practical 
account of preparing walls for painting in the manner 
of the ancients ; but they are content with three or four 
coats in all. 

" The lime used is a putty made from freshly burnt 
lime^ slaked in water, and allowed to rot for at least 
three months under cover in a damp cellar or disused 
well. "When it is first slaked it should be sifted into 
the water and carefully stirred every day for some time, 
so as to make it fine and silky. Lime slaked roughly, in 
lumps, does not get properly soaked and is troublesome 
to break up, and not being homogeneous spoils the work. 
It should be passed afterwards through a wire sieve and 
left as stated till required. The sand must be pit sand 
quite free from dirt or mud, and gritt5\ The marble 
dust is broken white marble pounded in a mortar till it 
is fine enough to pass through a piece of close muslin. 
The marble must be of a sharp quality, and not rusty 
from open-air sawing. Chips from a sculptor's studio 
are good for this purpose, but a magnet should be passed 
through them to collect all the particles of steel. 

^ opinions vary as to the superiority of chalk (fat) lime or girey 
limestone for this purpose. 



" If the wall of the building where the fresco is to go 
is of a porous, sandstone nature, it is well to have it 
cemented internally as a protection against damp 
drawing in from the outside. ^ From my own experiment 
in fresco the first coat was composed of brick and lime, 
the brick being pounded to about the coarseness of 
peas, well mixed with the lime, and laid on roughly 
about 1 in. thick, including thickness of laths between 
which it was finely keyed in. 

" The second and third coats of sand and lime, about 
the coarseness of brown sugar, were laid on about 
^ in. thick each, the first coat on the brick surface 
before the latter was dry, so as to amalgamate well 
together ; the second coat after the first sand one was 
well dried. This was of a finer texture than the first 
sand coat, and was finished off comparatively smoothly. 
These operations took about four months to complete, 
being carried out during the winter months. 

" The fourth coat of marble and lime is laid on by 
the fresco painter piecemeal as his work proceeds. 
* The intonaco is to be so prepared that it does not stick 
to the trowel, but easily comes away from the iron.' 
Sticking shows that it has not been sufficiently worked 
or beaten ; it should be pounded in a mortar (both lime 
and marble dust should have been previously strained 
separately), ground on slab, and, if at all uneven, 
crushed several times under edge of a strong palette- 
knife, which is the best method of detecting any dirt 
or grit. The proportions of the mixture of lime and 
marble dust should be equal by measure. It is fit for 
use when it is of the consistency of a thick paste. The 
best plan is to prepare a good quantity and place it in 

^ This was done to the wall of the school building upon which I am 
engaged at present and laths were stretched about two inches away 
from the wall to make the foundation upon which to place the layers 
of plaster. 



a cool earthenware pot with a Httle water. This last 
coat should be laid on as thickly as possible, though I 
found that ^ in. was the limit of adhesion. In greater 
bulk the intonaco tended to crack and eventually 
peel off. Probably the thickness depends greath^ on 
the greater or less smoothness of the sand coat beneath. 
Should any signs of cracks appear after la}dng on, tap 
the surface to find out by the hollow sound what is the 
condition of adhesion, and if unsound, cut out at once 
and relay. The presence of grease on the under-surface 
will prevent adhesion. The intonaco must be laid from 
the top, downwards, so as to avoid splashing finished 
work, which would inevitably happen if the lower 
part were completed first. 

" Sufficient for the day's work should be placed on 
a board. Always remember to have the tools which 
come in contact with the plaster well moistened ; this 
prevents sticking, and keeps the plaster compact. The 
arricciato must, of course, be most thoroughly wetted ; 
it is advisable in warm weather to sprinkle the portion 
of wall over night and keep it moist with wetted rags. 

" To lay on the intonaco satisfactorily it should be 
applied with steady pressure, spreading with the steel 
float and working from below upwards in the first 
laying on. After the principal mass of plaster has been 
applied the float can be used in all directions, so as to 
ensure its being even all over ; it must amalgamate 
with the under-coat, or it will show cracks and eventually 
peel off. I find floats of two sizes (adapted to the hand 
of the worker) necessary to get an even surface. The 
wooded float is indispensable to scour the wet plaster, 
so to speak ; where a large area is being laid, a large 
float is necessary. 

" The plasterer's tools principalh'^ used for laying on 
the plaster are : (1) the float ; (2) the trowel (3) the 



mortar board. Both wood and steel floats are used 
according to the nature of the work, but the former are 
best where the suction is strong. They are flat slabs, 
varying from 11x4 to 8x3, with a bridge-handle 
behind. Flat steel trowels of various sizes are used for 
pointing and filling in joins and crevices between the 
work of different days, and to give a finer finish to the 
surface. The mortar board, a square slab of wood, is 
the plasterer's palette. There are a variety of small 
tools for finishing edges — the most useful are slightly 
bent and spoon-shaped. A straight edge, for testing 
flatness of surface, and a plumb-line complete the outfit 

" Painting in Fresco on the Prepared Walls 

" A great deal of judgment is required in planning out 
the daily portion of work. I have not practically found 
it possible to hide the joins between different days' 
work, where these joins occur on plain colour surfaces. 
To begin with, it is difficult to lay the two edges together 
exactly in the same place. In the second place, it seems 
impossible to prevent streaking the colour. To get as 
exact a join as possible I have found it best at the 
beginning of the day to lay the plaster beyond the actual 
space marked, and at the close to cut back with a sharp 
scalpel to the desired area, cutting the edge obliquely 
and as cleanly as possible. In very hot weather it is 
advisable to leave the extra edge of plaster till the next 
morning, and then cut away, simply as a means of pre- 
serving the edges moister than if they were exposed by 

" With regard to this difficulty, and I believe impossi- 
bility, of true joining in colour, it has to be realized that 
the same tone laid on in various coats lightens in value 
with each coat ; so that where the edge of the second 



day's work overlaps that of the first, streaks of a lighter 
value are apparent. Overlapping cannot be avoided, as 
the edge, requiring to be moistened, sucks in some of 
the fresh tone. The only solution of the difficulty lies in 
planning the joins to follow outlines in the design as 
far as possible, or wherever two colours or shades meet- 
ing make a natural variety in tone. This, of course, 
produces much longer lines of joins than if worked in 
rectangles, such an extensive boundary-line giving a 
sense of weakness. It also has the drawback of drying 
more quickly. However, I have foimd no ill effects 
result from these intricate junctures, and they most 
effectually conceal defects in plastering of edges, and 
in colour. The great drawback lies in the frequent 
difficulty of having to paint on separate days the object 
and its background. To obviate this it is often worth 
while putting in some of the background, in order to 
get the object in right relief, and next day cutting back 
to outline and replastering for the fresh work. 

" Use of Cartoons 

" Many directions are given by the old writers on the 
subject of the Preparation of Cartoons. 

" With regard to the piecemeal transfer of the cartoon 
to the wall, I have found it less destructive to the 
cartoon to trace each portion on tracing paper and so 
transfer to wall, using a blunt wooden point for indent- 
ing the line as less apt to tear the paper than a metal one. 
The countersigning is very important for exactness, as 
in large surfaces one is apt towards the end to get some 
inches out in the design. 

" Remember in planning out the transfer of design 
that the fresco must be worked horizontally, beginning 
at the top of wall surface. It can either be worked 



from right to left or the reverse, according to the liking 
of the artist ; but, as has been said already, it must 
always be worked downwards as the splash of the 
material spots whatever is below. 

" With regard to the material for the cartoon, I find 
most congenial to myself the Willesden waterproof 
paper prepared with a thin layer of gesso ground. This 
is convenient for large surfaces as it is wide, and is sold 
by the yard. It is not so woolly as most brown paper 
and stands transport well. The gesso ground is not 
thick enough to flake off seriously when rolled, and it 
receives both charcoal and brush work in egg tempera 
admirably. The colour also resembles that of a wall more 
closely than white paper, and is less dazzling to the eyes. 

" In preparing the small-sized study of composition 
I find it invaluable to keep it always to scale ; that is, 
fix the size of the study to be I, |, or xV of the final 
design, so as to carry out the perspective to scale in all 
its details. But I need not dv/ell longer on these points, 
as all artists are equally well aware of the importance 
of exact preliminary studies for composition. 

" The Colours for Fresco Painting 

" A very limited range of colour is certainly the best 
safeguard of decorative harmony, as, the larger the 
masses of coloured surfaces, the simpler should be the 
colour scheme. I leave this part of my subject, only 
saying a word or two about the body-colour, white, 
which is made of the slaked lime used in preparing the 
plaster, purified by boiling and grindmg and allowed 
to dry in the sun till it is in a caked condition. It is 
very fine and white, and is preferable to using the 
intonaco, which, containing marble dust, is sometimes 
gritty and leaves scratches on the coloured surface. 



" With regard to the actual laying on of the colours, 
I am able to draw more upon my own experience. It 
is important to realize that fresco is practically water 
colour ; a stain rather than a paint, and it should be 
treated as a liquid body. The moist, ivory-like surface 
of the intonaco absorbs the colour, causing it to spread 
in soft gradations which no retouching can improve. The 
more transparent the pigment the better it is adapted to 
this surface ; and it is just this purity which gives such a 
brilliant softness to draperies whose folds are dra^vn in 
richer shades of the local colour of the stuff. It is for 
this reason also, I think, that Cennino's account of his 
master's method of painting flesh is so invaluable. 
Having first shown how to draw and model the outlines 
in some monotone, as a basis for the colouring, he says : 
' Procure three small vases and make three shades of 
flesh colour ; that is, the darkest, and the other two 
each lighter than the other in regular gradation. Now 
take some colour from the little vase containing the 
lightest tint and with a very soft pencil of bristles, 
without a point, paint in the lights of the face ; then 
take the middle tints of the flesh colour and paint the 
middle tints of the face, hands, and body, when you 
paint a naked figure. Afterwards take the third vase 
of flesh colour and go to the edges of the shadows ; and 
in this manner, softening one tint into the other, until 
it is all covered as well and as evenly as the nature of 
the work will permit. But if you would have your work 
appear very brilliant, be careful to keep each tint of 
flesh colour in its place, and do not mix one with 

" This is practically painting au premier coup, and 
I have tested the tnith of it by having previously 

1 The correct translation of the last clause is " except that with 
skill you soften one delicately into the other." B 


Example of fresco painting from the Church of S. Maria Xovella, Florence, 
by Ghirlandaio 


worked on other lines. Of these other methods I must 
mention one to caution you against it, as in theory it 
sounds so very attractive and is also recommended by 
an old writer (Armenino). First (just) let us suppose 
that a head has to be painted. The modelling is put in 
in semi-tones, which allows more time for intricate play 
of light and shade than in Giotto's method ; then the 
whole is pulled together by a wash of flesh tone laid 
lightly over the surface. This looks fresh and forcible 
whilst wet, but as it dries grows spotty and mealy- 
looking, till at last there is scarcely a true colour left. 
This is due to what I have already referred to as being 
of such vital importance to realize, namely, that fresco 
colours lose value in drying in proportion to the number 
of washes laid on. 

" On that account in work that is modelled by 
hatching, the chief difficulty is to hit upon the right 
hatching tone ; it must be darker on the palette than 
the pure shadow tone, or, when laid in place, will become 
either similar in value to the shadow or lighter. Much 
of the chalky or troubled look of fresco is due to this ; 
the shadows look hollow with cold reflected lights, 
because the hatching strokes dry lighter than shadow 
wash. It is a mistake to think that fresco must be 
chalky ; the range of tones is cooler than in oils, but 
not less pure, and they should not be opaque in quality 
any more than in water colours. It is for this reason 
that I do not understand the use of lime-water in pre- 
paring colours being recommended by some old WTiters 
in place of pure water. I have followed Professor 
Church's advice to employ baryta water in moistening 
the siu-face of the intonaco before laying on the colours, 
to prevent efflorescence ; but have given up adding it 
to the pigments, or moistening any coloured surface 
with it on account of its clouding water as lime does, 
o 209 


I would advise anyone who is interested in this to con- 
sult Professor Church's Chemistry of Paints and Painting. 
" In fresco, work for as direct and natural effects as 
possible ; you may rely on its drying harmoniously and, 
as it is called, decoratively. It takes a great deal to 
tire out the intonaco if it has been properly prepared ; 
should it begin to work up rough under your brush, it 
is only necessary to smooth it down with a wet trowel, 
being careful not to spread any of the plaster over the 
edges of the part in question. The best spirit in which 
to attack it is in that of the line of poetry : — 

' A wet sheet and a flowing sea, and a wind tliat follows fast.' 

" Difficulty of Correction 

" There is one more point in the actual handling of 
fresco which it is important to consider, indeed it is, 
I believe, one of its chief bugbears, namely, the difficulty 
of correction. There is really no great difficulty in it ; 
it is rather a question of patience. So long as there is 
plenty of time and daylight ahead of you, it is easy 
enough to correct on the plaster by going over the 
mistakes with a wet trowel and a little fresh intonaco ; 
but it is difficult if the day is drawing to a close or the 
mistakes are not noticed till the next morning. Then 
the only remedy is to cut out the offending piece, 
following outlines as much as possible. This is easily 
done, as the previous day's intonaco cuts like soap ; use 
a sharp scalpel, as thin-edged as possible, and cut 
slanting to make a better join eventually. Remove the 
old intonaco with trowel or palette-knife (or for a large 
surface a carpenter's scraper is most useful), wet the 
surface carefully so as to avoid splashing the surround- 
ing decoration, and lay on a fresh coat, joining 
thoroughly with the previous work. It is better worth 


O ^ ; 

7 —~. '^ - 

5 S - ? 

o v*. -* - 
~5- <^ 5 

< < Ti ■:* 
2 ft -■ 


correcting in this manner immediately or the day 
following, than waiting to see the effect when dry, 
because to postpone means that your intonaco is harder 
to remove — the portion to patch will by that time have 
been surrounded by work liable to be easily damaged — 
and it will be much more difficult to match the neigh- 
bouring tones. 

" It will be perceived that the actual fresco painting 
is not a lengthy process. The area one can cover satis- 
factorily depends, to a great extent, on the strength of 
the worker, also enormously on the importance of 
subject treated ; a head which covers, say, 1 ft. square, 
will probably take as long as 3 ft. square of background ; 
it also depends on the distance of the decoration from 
the eye of the spectator. Work at a height of 20 ft. 
from the ground requires free strong handling, which 
would appear insufferably coarse if on a level with the 
eye. There is no standard, though I think that a care- 
fully finished life-sized head in one day's work is good 
average speed. The great thing to my mind to avoid 
is, accepting the rate of oil painting as a standard of 
speed. Fresco must be fresh and spontaneous. You 
must not ignore all the previous time in preparation 
of cartoons and drawings. Nevertheless, the novice is 
apt to be disheartened by the coarse look of the work, 
as seen close at hand on the scaffold, but this essential 
feature of the work must be accepted. One must be 
willing to sacrifice personal gratification in a beautifully 
finished portion for the general effect of the whole. 

" With regard to style and the lading on of colours, 
the more individual the style the more likely fresco 
lb to live again. Mannerisms should be avoided, such 
as that of the inevitable iron-drawn outline, probably 
only a tradition of renovators ; or the return to archaic 
forms which were simply a primitive style of drawing. 



Fresco is so essentially plein-air in its effects that it 
should be the medium in which modern thought in Art 
should express itself best. 

" Preparation of Colours 

" We have now arrived at the last division of our 
subject ; it is fully as important technically as the 
preparation of the wall ; namely, the preparation of 
the colours. To anyone accustomed to palette and tubes 
of paint, the question of colours will not appear so very 
vital, but you who are accustomed to preparing powder 
colours for tempera work will understand if I go into 
some detail on the point. 

" In the first place, we must make a small study in 
colour of the design, yet large enough to show grada- 
tions of tone. In preparing the design, keep in view 
the object for which it is being made, so that the scheme 
shall be suited to fresco, not to oil painting. Legros' 
favourite saying, ' la belle et sainte simplicite,^ ought to 
reign here if anywhere ; and as much plein-air effect of 
light as is compatible with the subject. Then, with the 
study hung up over the grinding slab and a full range 
of powder colours in glass bottles (so as to recognize each 
at a glance) within reach, set to work to determine what 
shades will be required, distinguishing between those 
which can be produced by more or less thinning and 
blending on the wall surface, and those which have 
their own individuality. I have found them divide into 
two classes, namely, the pure or unmixed colours, such 
as black, Venetian, light and Indian reds, ochres, etc., 
which are indispensable in each day's paintmg ; of 
these, having always a supply in your bottles, you have 
a standard of colour to which they are bound to return 
when they are dry. The second class contains the 
mixed colours. These will, of course, produce unkno^vn 



tones when dry if you have only seen them mixed in 
the moist condition. 

" This brings us face to face with one of the chief 
difficulties of fresco — the alteration of tone after it has 
been laid on the plaster and become dry. To the old 
Italian masters who had been trained to it from their 
youth and lived in the midst of tradition, this was not 
the difficulty it is to us ; but still, it is mentioned as 
a serious hindrance. They used the pietra ombra as 
the modern Italian Avorkmen still do ; that is, a porous 
stone, which, absorbing moisture quickly, soon dries 
colours laid upon it and discloses its ultimate shade. 
But in England this stone is difficult to get, and it is 
besides a clumsy method where many shades are in use. 

"It is also possible to have a comparative scale of 
values showing the same colour in its moist and, side 
by side, in its dry state ; but this is only approximate. 

" I have employed another method suggested to me 
by a friend, accustomed to work in pottery, which is so 
simple, and I, at least, have found so satisfactory, that 
I am sure you will be glad to know of it. It is just 
simply to mix your powder colours dry, to the shade you 
wish to have. By very intimately blending the powders 
on the slab (being extremely careful that all the tools 
and the slab itself are absolutely dry) with a glass 
muller ; alternately crushing with the edge of the 
palette-knife, which is the best means of detecting 
impurities ; the different pigments used combine so 
closely that one can rely on the tone produced to remain 
true to itself, after being moistened, spread on the wall, 
and eventually dried out. Of course, this does not apply 
to artificial pigments, which, as shown above, can under 
no conditions keep their colour. In order to ensure 
matching work done on different days and in varying 
lights, compound a fair quantity of each shade, and 



when quite satisfied that the tint has been mixed by 
mechanical combination as completely as possible, cork 
up in a clean test-tube with name of colour, and which 
shade it represents, with a note as to constituent pig- 
ments ^mtten on a strip of tracing paper gummed 
round the test-tube. Do not grudge devoting several 
days at the beginning of your work to this preparation. 
It will be time gained in the end if only that it has 
made you so acquainted with j^our colour scheme that 
you feel master of it when the time arrives for attacking 
it on a larger scale on the wall. 

" Having got your stock of pigments in good order, 
it only remains to accustom yourself to a method for 
the day's work. The most excellent plan is to lay them 
out overnight, moistening the pans in which they stand, 
which should be well covered over with an oiled sheet 
of paper to prevent evaporation. It is always a pity 
to have to spend the fresh morning hours in mechanical 
grinding on the slab. 

" First decide on your colour study, how large an 
area you are likely to cover in the day's work, and 
what colours will be required and approximately how 
much of each. Then take from each test-tube, in turn, 
some of its contents and grind with pure water on the 
marble slab, beginning with those lightest in tone, as the 
slab and muller are apt to get stained by each pigment 
in spite of constant washing. Lay each colour when 
ground to a paste in a little white saucer, and set the 
saucers side by side in any regular order you prefer, 
on a slab of wood or board of white wood, as you can 
then paint a streak of each colour close to its own 
saucer on the board, which drying, will be an index 
to the eventual tone of the moist pigment. A pipkin of 
clean water in the middle of the board completes an 
extempore palette, which placed on a high stool forms 



a kind of table. The working palette on which the shades 
can be mixed is very good, made of aluminium with 
depressions to hold extra large quantities of paint ; it 
is of the ordinary shape and is held on the thumb, com- 
pleting with the brushes, trowel, and palette-knife the 
equipment of the artist." 

A method which falls to be discussed here, and of 
which examples are to be seen in the House of Lords, 
is the use of silicate of potash as a binding material ; it 
is known as the stereo-chrome process, and it was con- 
sidered to be absolutely permanent. " The Meeting of 
Wellington and Bliicher after Waterloo," and " The 
Death of Nelson," by Maclise, and " Moses Bringing the 
Tables of the Law," by Herbert, all in the House of 
Lords, were executed in this process. Within ten years 
Sir Arthur Church had to treat the decaying surface 
with paraffin wax. The process was afterwards improved 
by Keim. 

The process consists essentially of painting on a dry 
surface with the pigments mixed with water, and then 
fixing with a solution of silicate of potash. The ground 
is also soaked with fluo-silicate before painting, the 
object being to try to convert the lime into insoluble 
fluo-sihcate to form the binding material. The details 
are very complicated, and the pigments and solution 
require to be specially prepared. An example of the 
method in its final form, as improved by Keim, is to be 
seen in the fresco painted by Mrs. Lee Merritt at 
Wonersh, near Guildford. 

While the method probably results in a more durable 
painted surface than Buon-Fresco under the conditions 
of our polluted air, it cannot be claimed as a solution of 
the problem of permanent wall decoration. A very large 
portion of the lime and marble dust remains unchanged 



and will readily be attacked by sulphur acids, and with 
the formation and crystallization of sulphate of lime, it 
^vill break up the surface. For a full description of the 
process, see Papers read at the Society of Arts b}'^ the 
Rev. J. A. Rivington in 1884, and by Sir Robert Austen. 
The process is also described in Church's Chemistry of 
Paints and Painting, and in Mural Painting, by Hamilton 

In conclusion, the objection of chemists to the use of 
Buon-Fresco is due to the pollution of the air in this 
country with sulphur acids from the burning of the 
sulphur in coal. These acids are distributed in the air 
over the whole country, attacking and destroying lime- 
stone buildings in remote districts. The conversion of 
carbonate into sulphate of lime is not always equally 
destructive, and seems to depend on the aggregation and 
gro^vth of the crystals in certain centres. When fairly 
equally distributed, a limestone can carry from 4-5 per 
cent of sulphate of lime with impunity. The conditions 
governing the distribution are very obscure, and it might 
well happen therefore that rapid decay might take place 
in one case by the sulphate of lime crystallizing in definite 
spots and breaking up the surface ; while in another case 
this local crystallization not taking place, the frescoes 
might last longer. Generally speaking, the danger of 
such local crystallization is very much diminished if the 
wall and building are dry. 


Example of fresco painting by Beuozzo Gozz )li in the Chapel of the Eiccardi, Florence. 



We shall next proceed to consider other methods of wall 
painting. Before doing so, a word or two is necessary 
on the aesthetic as well as the chemical aspects of the 

If the frescoes by Giotto and his disciples are examined, 
the decorative effect produced is essentially a part of the 
wall- sm'face and is in no sense a picture painted on a 
wall. This is not entirely due to the method of painting 
adopted. The frescoes in the Church of Santa Croce in 
Florence are painted on a ground tone which matches 
the peculiar greyish yellow colour of the Florentine stone 
of which the church is built, and lights and shadows are 
a little above and below this, the colour being in the 
nature of a soft tinting of this fundamental tone. The 
whole effect is as if the stone itself had come to life, as 
if the natural varieties of tint of a stone had expressed 
itself in form and colour. Doubtless these frescoes have 
diminished in brightness, but I must believe that Giotto 
painted with this decorative object. The effect is en- 
hanced by the fact that his treatment in the drawing, 
and his decorative conception, is the same as the sculp- 
ture done under his direction. Painting as an indi\'idual 
art with its own methods of expression had not asserted 
itself. When we compare with Giotto's work the some- 
what garish frescoes in the Library of the Duomo at 
Siena, this conception of wall decoration has been de- 
parted from, the free introduction of gold leaf making 



this departure obvious. The question naturally arises, 
how far the decorative aim of Giotto depends upon 
the skill of the artist rather than upon the method 

Treatment mth paraffin wax melted in, as carried out 
by Sir Arthur Church in the House of Lords, is, I admit, 
fatal. At certain angles the frescoes are invisible ; but 
flat painting in oil, even executed on canvas attached 
to the wall, is capable, I believe, of successful treatment, 
and in this direction I wish to direct artists to the possi- 
bilities of painting in Chinese wood or Tung oil. This 
oil, the drying oil of China, dries flat without any treat- 
ment with wax and is one of the most durable of drying 
oils. It also makes excellent emulsions, with size and 
water. Like all drying oils, it darkens with time, but 
if thinly painted on a pure white ground this danger is 
reduced to a minimum. 

In order to be able to execute permanent wall painting 
in our modern cities, it is essential that the ground on 
which the fresco is painted must be secure from attack 
by the sulphur acids in the air. This is easily achieved 
by replacing lime plaster by plaster of Paris. Plaster of 
Paris alone produces a cold and uninteresting surface 
and is apt to crack ; but plaster of Paris with white sand 
produces an excellent painting surface. This should be 
made non-absorbent with size as directed by Cennino 
Cennini. The size, to protect it from change, should be 
lightly sprayed over with a dilute solution of formalin. 
On this the artist can paint with Tung oil in which the 
colours have been stiffly ground, thinned vnth a little 

There is another direction in which experiments re- 
quire to be made, namely, the use of silicon ester as a 
fixative. This solution, which quickly evaporates after 
application, finally, by the action of the moisture in the 



air, deposits a layer of hydrated silica which is in itself 
indestructible ; and the process has the advantage over 
methods depending on the use of soluble silicates that 
no soluble salts are formed at the same time, ■which may 
cause further trouble. It should prove very valuable to 
save and protect existing frescoes and remains of old 
colouring in our churches. 

I shall now briefly describe other materials for wall 

On the whole, the evidence is in favour of the use of 
size in many cases for mediaeval wall decorations. If 
size is used, the finished fresco should be sprayed with 
formalin. Casein dissolved in soda, or ammonia, or mixed 
with lime, can also be used, and Professor Oswald recom- 
mends pastel as a wall decoration, and fixing with casein. 
Emulsions can also be used. Of these, the best in my 
opinion is Chinese oil and size. A final spraying to fix 
the size with formalin is advisable. 

The Gambier Parry medium, as modified by Sir 
Arthur Church, is a mixture of paraffin wax, 4 oz. 
(M.P. 58° — 62° C.) copal oil varnish, 16 fluid ozs., and 
oil of spike, 12 fluid ozs. This medium was used by 
Madox Brown for the decoration of the Town Hall in 
INIanchester. Half of the paintings are on plaster. 
They have stood well, but here and there the sulphur 
acids are attacking the plaster below, and crystals 
of sulphate of lime are forming and breaking up the 
paint in places where the paint is very thin. The 
other half are painted on canvas, and are on the whole 
in good condition ; but in some cases the paint is lea\ing 
the priming so that whole sheets could be ripped off with 
a knife blade inserted underneath. This seems to be a 
vice peculiar to copal varnish in emulsions. The frescoes 
in this medium on canvas in the McEwan Hall, 
Edinburgh, which are now some thirty years old, seem 



in excellent condition. On the other hand, Lord 
Leighton's fresco in this medium at South Kensington 
was in a deplorable condition, but has been successfully 

This medium has two defects. It is apt to attract the 
growlh of mould. This takes place on the organic pig- 
ments, like madder lake and Prussian blue, which should 
therefore be excluded. If the attack is superficial, it can 
be cured by several sprayings at intervals with formalin. 
If deep-seated, nothing can remove it, not even scraping 
off the paint and repainting. A damp wall or a cold wall, 
causing condensation of moisture, is the cause of this. 
For this reason an inner wall should be built, ventilated 
from inside the building. 

If the fresco is painted on canvas which is to be 
attached to the wall by the usual process of cementing 
with stiff white lead, the canvas should be sprayed with 
formalin and then washed over the back with a very 
weak solution of corrosive sublimate. 

Another objection to the medium is that, o\\ang to 
its complex nature, it is difficult to clean it safely. 
Fuller's earth, breadcrumbs, or saponin and water are 
tlie safest. 

It is evident that under the present unfortunate con- 
ditions of the air in our big cities it is very difficult to 
give a positive ruling as to the method of wall decoration 
which is likely to be most successful. The reasons 
against Buon-Fresco are, I think, sufficiently conclusive ; 
and the Gambier Parry medium has obvious defects. 
Probably, therefore, the safest thing to do is to accept 
the necessity of, at any rate, an egg-shell gloss and to 
paint very thinly in oil, preferably in Chinese wood oil 
instead of linseed oil, and to protect the surface by the 
double treatment used by Mrs. Traquair. It is also 
evident that where the use of lime is not necessitated 



by the process adopted — namely, Buon-Fresco — the 
painting should be executed on a surface of plaster of 
Paris and sand. 

It is of some interest at this point to consider what the 
Northern tradition was in the matter of painting on walls. 
Mr. Noel Heaton found on certain remains of Roman 
villas in this country that, in order to carry out a fresh 
painting, a thin coat of plaster had been laid on the older 
painting, which makes it probable that the Roman 
method was Buon-Fresco. In some samples which I 
examined from the foundations of the new Post Office 
in London, in which the plaster was sound and the pig- 
ment firmly adhering, I could find no indications of any 
organic medium, which is a further confirmation of the 
view that the Roman method was some modification of 
Buon-Fresco. Mr. Cran, while of the opinion that size 
was largely used in England for wall decoration in the 
Middle Ages, mentions a similar case of new plaster laid 
over an old painting at Chichester, which again suggests 

That size was probably often used for temporary 
decorations of no great importance is quite probable, 
but it can hardly have been used for permanent and 
important works. 

We possess in England a great number of accounts for 
painting of the thirteenth and fourteenth centuries. In 
the account rolls of Henry III, Edward I, and Edward II, 
oil is mentioned, and again in the accounts of Ely 
Cathedral from 1325 to 1352. The account for the 
decoration of St. Stephen's Chapel from 1352 to 1358, 
and in the account of Edward I, 1274 to 1277, relating 
to the painted chamber, white lead, oil, and varnish are 
mentioned. The varnish is sold by the pound, and may 
well therefore be, as Eastlake suggests, the resin, possibly 
sandarac, which was purchased and dissolved in oil to 



make the varnish. Such a varnish would be very slow 
drying and explain the purchase of charcoal for drjdng 
the painting. ]\Ir. "Wilson, of the Office of Works, has 
WTitten an interesting paper on the Westminster Ac- 
counts. The blues are indigo and azurite. This use 
of azurite is of interest, as the blue in illuminated manu- 
scripts, with certain exceptions during the thirteenth 
and fourteenth centuries, is real ultramarine. I have 
found azurite only on certain Flemish manuscripts of 
this period. Evidently azurite was the common blue 
at this period of time, although not considered good 
enough for illuminated manuscripts. The large quan- 
tities of oil used seems to point to extensive wall surfaces 
ha\-ing been covered with oil painting. That oil was not 
the universal vehicle in England is shown by another 
painting account, quoted by Eastlake, of 1274 to 1277, 
where eggs, honey, white wine, and size are the vehicles. 
Size is occasionally mentioned in the other accounts, but 
in very small quantities. The question arises whether 
the walls were sized and then painted in oil and varnish, 
or painted in size and then varnished, or whether oil 
alone was used both to prepare the ground and to paint 

Cennino Cennini directs to size first and then paint in 
oil, but the practice of house-painters when painting on 
stone outside a building is to saturate the stone with oil 
first ; and 3Irs. Traquair has adopted this method in her 
frescoes, presently to be described. 

Judging by the small quantities of size mentioned and 
the large quantities of oil, it is most probable that the 
plaster was first saturated Anth oil and white lead ground 
in oil ; that then the painting was continued in oil, and 
was finally coated -with varnish which required to be 
heated with charcoal as it Avas so slow in drying. As I 
have shoAMi in the chapter on Oil, the mediaeval recipes 



would give an oil sufficiently quick in drying, and the 
charcoal would therefore only be required for drying the 

The whole technique, judging by the accounts, is the 
same as that adopted by ]\Irs. Traquair for her frescoes 
in Edinburgh. 

The cleaning of the painted tombs in Westminster has 
been carried out recently by Mr. Tristam, who is of 
opinion that the paintings were executed in size and 
then varnished with oil. 

As there has been some discussion as to whether the 
wall paintings in this country were executed in oil or in 
size and afterwards varnished, it is of some interest to 
quote the following recipe from the Strasburg Manu- 
script : — 

" I have now honestly, and to the attentive, amply 
taught how all colours are to be tempered, according to 
the Greek practice, with two aqueous vehicles ; also 
how the colours are to be mixed, and how each colour 
is to be shadowed ; (I have told) the whole truth. I 
will now teach how all colours may be tempered with 
size, on wood, on walls, or on cloth ; and in the first 
place, how the size is to be prepared for the purpose, so 
that it shall keep without spoiling, also without an un- 
pleasant smell. Take parchment cuttings, and, after 
washing them well, boil them in water to a clear size, 
neither too strong nor too weak. When the size is suffi- 
ciently boiled, add to it a basinful of \inegar, and let the 
whole boil well. Then take it from the fire, strain it 
through a cloth into a clean earthenware vessel, and let 
it cool. Thus prepared, it keeps fresh and good for a 
long time. The size being like a jelly, when you wish 
to temper any colours, take as much size as you please, 
and an equal quantity of water ; mix the size and water 



together, and likewise much honey with them. Warm 
the composition a Uttle, and mix the honey thoroughly 
with the size. With this vehicle all colours are to be 
tempered, neither too thickly nor too thinly, like the 
other pigments of which I have already spoken. And 
these colours can all be coated with varnish ; thus they 
become glossy, and no water nor rain can then injure 
them, so as to cause them to lose either their tints or 
their shining appearance." 

From this it is evident that painting in size and then 
varnishing was well known. 

I obtained some tiny samples from the tomb at West- 
minster and found gold, vermilion, and azurite for the 
pigments, and was satisfied as to the presence of oil, 
though whether used in the way suggested by Mr. 
Tristram I could not say. Most of the azurite had turned 
quite black, just a few blue crystals still existing here 
and there. Whether this blackening was due to the con- 
version of the azurite into sulphide or oxide I do not 
know, as I had not sufficient for further experiments. 
Under the azurite a curious technique had been adopted : 
first a layer of white lead in oil next the stone, then a 
gesso containing fine fragments of fibre, and then the 
azurite in oil. 

No such elaborate preparation is found under the 

As far as I am aware, no chemical examination has 
been made of the vehicle of the various remains of 
frescoes throughout England, so that any opinion as to 
the medium is guesswork. On the whole, the evidence 
from the numerous accounts is in favour of the view that 
wall painting in this country was executed in oil and 
varnish, thus producing a shiny surface, and that there- 
fore, in trying to transplant Buon-Fresco to this country, 



the modern artist is going directly against the Northern 

As a modern example of the use of oil and oil varnish 
for wall decoration, the frescoes carried on in Edinburgh 
in various public buildings some thirty years ago by 
Mrs. Traquair will now be described. 

Among other places decorated by ]\Irs. Traquair is the 
Song Room of St. Mary's Cathedral. This is a stone 
building, lined inside with ordinary hme plaster, which 
has been laid directly on the stone, and the decorations 
were carried out some thirty-five years ago. 

In the first place, five coats of zinc white in oil were 
laid upon the plaster without any pre\^ous sizing. The 
first coat was very thin, being highly diluted with tur- 
pentine. Each coat was allowed a week or two to dry, 
and then followed by another coat containing a little less 
turpentine, the final or fifth coat consisting of almost 
pure zinc white in oil. Six weeks were allowed for this 
final coat to dry. The frescoes were painted with ordi- 
nary oil colours, thinned down mth turpentine, a piece 
of beeswax about the size of a hazel nut being dissolved 
in each half-pint of turpentine. The pigments used were 
light red, burnt sienna, yellow ochre, terre verte, cobalt 
blue, real ultramarine, and a very little Chinese ver- 
milion, which when glazed ^\ith light red gives a rich 
and beautiful colour. No black, white, or bro^\Ti was 
used, the whole of the light being obtained by the 
thin transparent washes of oil colour on the white 

In this matter Mrs. Traquair's technique is the same 
as that adopted in Buon-Fresco, and is of interest in 
connection with the whole theory of colour which I have 
developed in this book. 

When the painting was dry it was varnished with an 
old carriage varnish, thinned down ^nth turpentine. 

P 225 


Over this a flat varnish was laid, and this again, when 
dry, was poUshed mth beeswax and turpentine. 

The whole fresco having got very dirty, Mrs. Traquair 
washed the whole of it a short time ago with soft soap 
and w^ater, and then repolished the beeswax with the 
palm of the hand. Here and there a crack has appeared 
in the plaster, and in one small portion some early 
cracking took place, but this cracking has gone no 

In one place over a window, where the wall is very 
thin and exposed to the full blast of the east wind, which 
brings with it all the worst weather in Edinburgh, owing 
to damp getting through, the plaster became detached 
from the stone, but the fresco remained firmly attached 
to the plaster. These frescoes have now got a mellow 
golden tint ; the surface has an egg-shell gloss, and the 
whole effect is very beautiful and satisfying. As Mrs. 
Traquair says, it is useless in our modern cities to do any 
wall painting which cannot be thoroughly well washed 
down, and for this a smooth surface, thoroughly pro- 
tected by oil and varnish, is necessary. 

I cannot help believing that in these frescoes Mrs. 
Traquair has got much nearer to the mediseval Northern 
technique than most modern efforts. 




It is obvious from what has gone before that a picture 
is a somewhat complex combination of substances with 
different properties, and there are so many different 
varieties of grounds, methods of painting, and varnish- 
ing, that no one picture is exactly like another, and 
consequently their preservation and cleaning and 
restoring is no easy matter, and only broad general 
instructions are possible. 

For the art gallery itself certain general rules should 
be followed. Freedom from injurious dust and soot 
particles is evidently, from what we have learnt, essen- 
tial. The soot of our big cities has a most corroding 
property. If a surface of oil paint on a picture which 
has been exposed to the London atmosphere for years 
^^ithout protection is examined under the microscope 
it is seen to be covered with little black particles. Wash- 
ing with, turpentine, or even with soap and water, re- 
moves some superficial dirt, but does not remove the 
ingrained tarry particles which seem to have eaten their 
way into the oil surface. They can only be removed 
with a solvent which at the same time removes some of 
the paint of the picture. If the picture has not been 
protected by a varnish, injury to the picture itself is 
ine\'itable if an attempt is made to clean it. 

If the varnish is easily removed \\'ithout touching the 
paint film, well and good. If, on the other hand, an oil 



varnish has been used, in trying to remove it by solvent, 
we may well, if great care is not taken, remove some of 
the paint as well. All pictures in oil and tempera should 
be protected by an easily removed varnish or protective 

Of these varnishes, mastic is the one universally used. 
I should prefer a protective coat still more easily re- 
moved if possible ; beeswax and turpentine has been 
suggested in place of mastic varnish. I find that a 
polished film of beeswax has only a tenth of the pro- 
tective value of a film of mastic varnish. For the 
tempera pictures, a varnish readily soluble in alcohol 
would be better, as the tempera medium is quite in- 
soluble in this solvent. 

Protection by a removable varnish, and in addition 
protection by glass, is essential. The enclosing the 
picture in a glass frame is not so simple a matter as it 
seems. We wish to exclude dirt, but at the same time 
it is not safe to attempt hermetically to seal the picture, 
as in that case, under certain conditions of temperature, 
the moisture which the whole fabric of the picture 
contains may condense on the surface, encoiu-aging the 
growth of mould spores. 

The introduction of a thin strip of velvet against 
Avhich the edge of the glass is pressed, with the usual 
method of pasting up the back with brown paper, is 
a sound practice. The picture can then breathe, and 
yet it would be dust-proof, the velvet acting as a dust 

In the case of pictures painted in tempera which show 
signs of crumbling, or oil pictures which seem to show 
signs of scaling off the ground, it is a good plan to arrange 
for a line of visible black just inside the glass along the 
bottom of the frame, so that any disintegrated particles 
which have fallen off can be seen. The attacks of the 



wood-beetle can often be first detected in this way, a 
little wood dust revealing their presence. 

Where pictures have been relined, the back may be 
regarded as well protected ; but in the case of those which 
have not been so treated, Professor Oswald has made 
the excellent suggestion of covering the back with tinfoil. 

This can be easily and cheaply done, but if the method 
of relining is adopted, which I shall describe presently, 
and which is the practice in the Amsterdam Museum, 
it seems to me better to reline a picture at once, whether 
ancient or modern, when purchased by the Gallery, 
rather than wait for injury to take place ; and I have 
suggested in another place the coating of the back of 
the pictures with a layer of beeswax and resin, covered 
with a protective canvas, which is practically the Dutch 
method of relining. 

Pictures, and more especially tempera pictures, are 
apt sometimes to show a tendency to scale. In the case 
of tempera pictures, a little size should be introduced 
with a hypodermic syringe under the scaling portion, 
which should then be gently pressed home. In the case 
of oil pictures, a little quick-drying linseed oil can be 
introduced in the same way. 

Brittle blisters can be softened with chloroform, or 
a very, very thin solution of collodion spread over them, 
which will both soften the blister and at the same time 
protect them with an elastic film. A little linseed oil 
can then be introduced under the blister with a hypo- 
dermic syringe and the blister worked down with the 
fingers steeped in oil. Finally, all superfluous oil must 
be removed. 

A tempera picture, if it has not been varnished, can 
be cleaned with alcohol ; the loose pieces cemented 
down with size, and then a thin coat of size laid over 
the whole. 



Some people have stated that to varnish a cracked 
tempera picture with mastic is dangerous, as the mastic 
soaking in underneath the tempera will loosen it and 
cause it to come off. The custom of the best picture 
restorers in Florence is to lay over the picture a thin 
coat of size. This probably is all that is necessary in 
Italy. In this country it would be better to follow 
up the thin coating of size when quite dry with a layer 
of mastic varnish. The size will prevent the mastic 
varnish soaking in and doing any damage. 

A similar objection has been raised to varnishing 
with mastic oil pictures which have become dry, the 
pigment no longer carrying sufficient oil to bind it 
together. In this case it would be safer to rub in gently 
a little oil, taking care that the minimum of oil is used ; 
give this some six months to dry and then varnish with 

It is evident from what has been said as to the 
properties of wood and canvas that not only should the 
temperature of the room be kept as even as possible, 
but the percentage of moisture in the air should also be 
controlled, wet and dry bulb thermometers being used 
to measure this percentage of moisture. The air should 
be neither too dry nor too moist, a mean percentage of 
moisture being attained as far as possible. The pictures 
should never be exposed to direct sunlight. 

The superficial cleaning of pictures which have not 
been varnished with mastic is probably best done by 
wiping with cotton-wool, moistened with well-rectified 
spirits of turpentine. In the case of a mastic-varnished 
picture, the surface can be dusted, but a solvent cannot 
very well be used which will not cause the mastic 
varnish to become sticky. 

There has been much discussion as to whether water 
should be used in the cleaning of pictures. If water is 


^.v per: 

the Director of the Haarlem Museioi 

By Fraxz Hals. 

(Municipal Picture Gallerr, Haarlem.) 

The three phctographs, this and the two following it. illustrate the cleaning of a picture by- 
Hals in the Haarlem ilu.'-euin by Mr. De Wild. This photograph was taken before cleaniEg, 

■ctor of the Haarlem Muse. 

Br Franz Hals. 

By te>-i)i,ssw>i oj ;,;,■ L 

(Municipal Picture Gallery, Haarlem.) 
Here the picture is partially cleaued. The cleaning gave rise to some controversy 

until the 
materials used for the cleaning of a second picture by the same methods had been examined. 


to be used, the cotton-wool should be nearly dry, and a 
little saponin can be mixed with the water. This is an 
excellent and harmless cleaning agent, being prepared 
from the bark of the soap tree. On no account should 
any kind of soap be used. The objection to the use of 
water is that soaking through cracks it may loosen the 
ground priming of the picture. If excessive water is 
avoided, and the picture placed in a warm dry place to 
dry, there should be no danger of this, but soap would 
remain in the fissures and would probabl}^ ultimately 
do harm. 

A picture which has become low in tone can often be 
very much improved by placing it in a window looking 
north for a few months. Care should be taken that the 
direct rays of the sun do not fall on it. 

We shall now discuss the cleaning of pictures from 
the picture restorer's point of view — that is, the actual 
removal of varnishes, which have become scaled, dark 
brown in colour, and full of particles of dirt. As has 
already been stated, an oil picture which has never 
been varnished and which has become ingrained with 
London soot, cannot be completely cleaned without 
injury to the paint. \Miere there is a protective varnish, 
the problem is not so insoluble. There are two types of 
varnishes — the spirit varnishes, in which the resin is 
dissolved in alcohol or oil of turpentine, and which are 
readily soluble in a similar medium ; and the oil 
varnishes, in which the resin has been dissolved in 
linseed oil, and which dry in the same way as a drying 
oil, forming a film at least as insoluble as the paint of the 
picture itself. These obviously form the most difficult 
problems for the cleaner of pictures. 

Mastic varnish is not only readily soluble, but can be 
removed by gentle friction. If the corner of the picture 
is very gently but persistently rubbed with the tips of 



the fingers, presently a white powder is formed of the 
disintegrated varnish, and it is possible, starting from 
this corner, to work over the whole picture and remove 
the mastic varnish. Sometimes it is necessary to crush 
into powder a little bead of mastic resin in order to 
start the process. The powdered resin having been 
blown off, the cleaning process can be completed by 
rubbing lightly over with a piece of cotton-wool soaked 
in turpentine. It has been objected that even this 
method may result in injury to the surface of the 
picture, and in the case of very impasto paintings, it 
would not enable us to get into the hollows, and in that 
case we must trust to the solvent. 

When cleaning a picture of a varnish other than 
mastic, the picture cleaner uses two liquids — one 
alcohol and the other turpentine ; the alcohol being 
used as a solvent, and the turpentine as a restrainer. 
He has a dish filled with each beside him, a plentiful 
supply of cotton-wool, and taking some cotton-wool in 
each hand, he dips the one into the alcohol and the other 
into the turpentine, coming over the surface which he 
is cleaning with the turpentine, followed with the 
alcohol, thus alternatively dissolving and restraining 
and watching carefully the process of his work, he aims 
at the removal of the old varnish without injury to the 
paint below. 

If the alcohol fails to dissolve the varnish, we must 
use more powerful solvents, such as amyl alcohol, 
chlorinated hydrocarbons, caustic potash, or ammonia. 
All these solvents will not only attack the varnish, but 
will also attack the paint layer, and therefore must be 
used with the greatest caution. 

In the case of a painting in which the brush work 
shows ridges and furrows, the cleaning presents great 
difficulties, as the tops of the ridges are apt to be 



Another portion of the same picture partially cleaned. 

An analysis of a second picture cleaned by this method proved that the wads of cotton-wool 

used contained nothing but old varnish. 


over-cleaned, while the furrows still contain old yellow 

These processes are in my opinion always dangerous, 
for the following reasons. In the first place turpentine 
is far from a perfect restrainer, it being quite possible 
to remove paint from the surface with cotton-wool 
steeped in turpentine. 

The danger of using powerful alkalies like caustic 
potash, which will form soaps with the dried oil film, is 
obvious ; and I believe it is never used to-day by careful 
picture cleaners, but has been replaced by ammonia. 

Ammonia is volatile, and therefore quickly evapor- 
ates ; but its volatile nature is a source of danger. 
Fragments of oil paint some twenty years old, enclosed 
in a covered vessel over ammonia liquor and exposed to 
the vapour of ammonia, are quickly softened right 
through, being converted into a soft sticky mass. On 
exposure to air the film hardens again, but must now 
consist largely of ammonia soaps and be no longer 

The vapour of ordinary alcohol and amyl alcohol also 
rapidly soften the film of dried oil. Therefore during 
treatment the film of dried oil must be absorbing these 
volatile substances with more or less injury to its 
stability. One paint layer may be more easily attacked 
than another, so that I have found after exposure to 
ammonia vapour a skin of paint floating on a semi- 
liquid mass of the priming underneath. 

The use of chlorinated hydrocarbons is also to be 
deprecated, as it is uncertain how far they are per- 
manently absorbed by the film, and their stability is 
doubtful, with the possible release of chlorine, which 
will have disastrous effects. 

The difficulty of cleansing a painted surface with 
ridges of paint are sufficiently obvious. I have examined 



such a surface under the microscope after the picture 
cleaner had been at work, to find that the ridges of white 
lead were all torn and ploughed, while varnish was still 
lurking in the hollows. 

In Florence the practice of the best picture cleaners 
is to rely very largely on the use of the knife. This is 
the only possible way of removing overpaintings in egg 
and is the safest way to remove old varnish. Among 
the delicate instruments of the surgeon, the necessary 
tools can be found. A skilled workman can remove thin 
layers of paint and varnish without any injury to the 
surface below. In this way all dangerous solvents are 

If solvents are to be used, brushes instead of the harsh 
and wholesale cotton-wool should be adopted. As I have 
said, turpentine is itself a solvent, and a much more 
perfect restrainer is castor oil. This oil mixes with 
alcohol, and some of the pure oil having been laid on 
the surface about to be cleaned, a mixture of the oil 
with alcohol can be painted into the oil on the surface 
with a fine brush, and this process repeated until the 
varnish is removed. By using this mixture of alcohol 
and oil, the absorption of the alcohol by the film is 
diminished, and the solvent action of the alcohol is 
completely under control. The objection to this method 
is the leaving in cracks of a non-drying oil. The oil 
must therefore be washed off the picture with turpentine 
at the finish. Only pure alcohol should be used. 

When the stronger solvent may be necessary, copaiba 
balsam is saponified with ammonia, by adding ammonia 
liquor drop by drop to the copaiba balsam, and shaking 
up until a clear solution with a very faint smell of 
ammonia is obtained. This is a powerful solvent, and 
will, of course, attack a paint layer if pushed too far ; 
but diminishes the danger of the absorption by the film 


■3 a 

/: S 3 = 

— i 5 ^ 

— _ Z. !» 


of the free ammonia. At the finish the copaiba balsam 
must be washed off with turpentine. 

These two recipes I owe to a well-known Dutch picture 
restorer, and I have tried them and found that they 
work excellently well. In careful hands, therefore, they 
seem fairly safe, but nothing can guard from the danger 
of overcleaning. 

The aim of the cleaner should be to leave a thin layer 
of the old varnish, both on ridges and in hollows. 
Plenty of time and infinite skill and patience are neces- 

Relining. Before cleaning a picture, if the paint 
shows signs of being detached from the canvas, it is 
relined. This does not mean the removal of the old 
canvas, which is seldom done ; but backing the old 
canvas with a new one, which is cemented on, the aim 
being to drive the cementing material through the old 
canvas to the back of the paint layer. In order to do 
this, sheet upon sheet of fine tissue paper is pasted on 
the front of the picture to protect it from injury, and 
the canvas is cemented on to new stretched canvas by 
ironing the back with hot irons. In this country a 
mixture of glue and resin is usually used, and very 
heavy irons, with the result that the brush work is some- 
times flattened and ruined. The following method used 
in the Dutch Galleries should always be adopted, as 
being much safer and more satisfactory : 

After a picture has been stretched on a frame and 
protected on the front by thin sheets of paper pasted 
on it by means of starch, the new canvas is applied to 
the back of the old one. The material that acts as the 
adhesive substance between the old and the new canvas 
is a mixture in equal parts of wax and resin, with the 
addition of a little oil of turpentine. I have already 
spoken of the high protective value of this mixture. 



The new canvas is flattened with electric flat irons, and 
every care is taken that these irons are not too hot. 
The flattening itself is done on a table covered with 
soft paper on sheets of quilt, so that the impasto may 
not be crushed, and may remain undamaged. If the 
impastoes are very big, a blanket is put on the table. 
One of the advantages of the wax and resin mixture is 
the fact that if small portions of this material appear 
after the relining on the painted side of the canvas, it 
can easily be removed by using turpentine. 

In the case of rents in the canvas, a fresh piece of 
canvas cut to fit the hole is carefully inserted before the 
relining is done. Afterwards it is filled up from the 
front with a little gesso. Where part of the painted 
surface has disappeared, the bare portion is filled in by 
stippling on colour ground in mastic varnish. It is 
usual before proceeding to this restoration to put a 
very thin film of mastic varnish over the whole 
picture. Finally, the picture is varnished with 

In this treatment of mastic there is possible danger. 
If the pigment on the picture is too arid after the cleaning 
process it may mean the replacement of oil by mastic 
varnish as a medium, which is not advisable. In such 
a case a little oil should be rubbed in with the fingers, 
care being taken that the very minimum quantity is 
used, wiping off all excess, and this should be given a 
few months to dry before the final restoring and varnish- 
ing is done. It has been suggested by some authorities 
that where such restoration work is necessary, care 
should be taken that it should be slightly below the 
level of the surface of the paint, so that it is always 
possible to detect where this has been done. It is un- 
necessary to say that all such restoration should be 
confined to places where actual gaps in the paint surface 


(Tate Gallery.) 

This photograph illustrates the damage done to impasto by the habit of the English reliners 

of pictures of using very heavy irons to drive the composition between the new and the old 

canvases through the old canvas. 

The method of relining adopted in the Continental galleries, which is described in the text 

from the account of the Director of the Rijks Museiun, is quite different, and preserves the 

impasto from injury. The Hittening of Turner's beautiful impasto is quite obvious in this 



occur. A skilful restorer can do this without any need 
to come over any of the painted surface. 

Finally, photographs should be taken both before 
and after cleaning and restoration, for the sake of 

The bad appearance of a picture is sometimes due 
to the surface disintegration of the varnish, and, if this 
is a spirit varnish, the exposure of the picture to alcohol 
vapour redissolves the varnish and restores the surface. 

This treatment of alcohol vapour is known as the 
Pettenkofer process, and has given rise to much dis- 
cussion and controversy. It can sometimes be used 
with benefit if applied by those skilled and experienced 
in the process. It is evident from what I have already 
said about the powerful action of alcohol vapour on 
the dried oil film that unless used with the utmost 
caution it may completely destroy the picture. 

One of the most difficult problems in the past has 
been how to destroy the wood beetle, when found to be 
attacking a panel, without doing any injury to the 
picture. It is necessary to find a vapour which, while 
destructive to the beetle, will not act on oil or varnish. 
Dr. Scott has solved this problem, as he finds the vapour 
of bisulphide of carbon can be safely used. To prevent 
a return of the beetle the holes should be filled up with 
beeswax mixed with sandalwood oil, as the female 
beetle dislikes the smell of the oil. The backs of panel 
pictures should occasionally be rubbed over with 
sandalwood oil for the same reason. 




It will be evident to a student reading this book that 
the craft of the painting of pictures has suffered a 
terrible loss from the perishing of studio traditions, handed 
down from age to age through the apprenticeship system. 

The loss of the tradition is not entirely to be ascribed 
to the gradual disappearance of the teaching of painting 
as a craft and the separation of the painter from the 
artist colourman. It is also due to the wonderful flexi- 
bility of the oil medium in fulfilling the needs of every 
ncAv type of artistic expression. 

The early methods of painting in oil were, as we have 
seen, scientifically sound ; but, being merely the result 
of tradition without any scientific basis, proved useless 
as a guide to the painter when he abandoned the early 
methods because they no longer enabled him to produce 
what he wished to express. 

I have dwelt at some length on what we know from 
MSS. as to the history of oil painting, as it seemed 
necessary to demonstrate that evidence is conclusive 
that we have no lost medium or special trick to look for 
in order to explain the durability of the early pictures. 

Having got rid of this search for the philosopher's 
stone we have found that an examination of the early 
methods in the light of the optical principles involved, 
is quite sufficient to account for the excellent results 
obtained. We have seen how the early oil painting is 
really a combination of tempera with oil, and how old 



traditions handed down enabled Van Eyck and his 
followers to adopt methods which we have found to be 
optically sound. 

Later, when the tempera under-painting is abandoned 
and the oil paint is laid direct on the gesso, we still find 
the same reliance on the white gesso to correct the 
defects developed by the oil in time ; in fact, the use in 
oil of a water-colour technique. When later the practice 
of solid white for high lights becomes more universal, 
great care is taken to grind the white stiffly with a 
minimum of oil. But with the replacement of the white 
gesso panel by canvas, this gesso was, in course of 
time, replaced by an oil priming, and as the fundamental 
optical principles involved were not understood, the 
methods were not adapted to the new conditions. But 
even here we find that all through the history of painting, 
including many of our modern painters, sound optical 
principles have been followed, more perhaps by instinct 
than by reason. 

In this gradual losing of sound tradition and sound 
scientific principles, the man of science has perhaps 
not been without blame. The variety of new pigments 
invented owing to the rapid development of chemistry 
in the last hundred and fifty years, naturally made him 
turn his attention to the examination of these pigments, 
in order to discover which were permanent and which 
fugitive, and the results of his experiments have been 
applied by the painter in oil without full understanding. 
The wrong and right uses of such pigments as cobalt 
blue when ground in oil is a striking instance. 

I therefore attempt in this book to emphasize the other 
considerations due to changes not in the pigment, but 
in the oil itself, which must guide the painter in oils, 
if he is to produce a picture which will not rapidly lower 
in tone. 



Of modern painters in oil, the Pre-Raphaelites alone 
seem to have thought out and adopted a definite tech- 
nique with a view to obtaining permanent results, a 
technique, as we have seen, sound in principle but 
limited in power of expression. It is here that the 
difficulty lies to keep sound scientific principles in view, 
while obtaining full freedom of expression. The study 
of the problem seems to have been abandoned in despair 
by the painter in oil, and while the modern painters in 
tempera have been striving to be, above all, craftsmen 
with an accurate knowledge of the properties of their 
materials, the painter in oil seems, in too many cases, 
to have given up any attempt to master the scientific 
properties of his materials. 

I have before me, as I write, a landscape painted by 
a famous French artist, in which he has deliberately 
prepared a priming of asphaltum on which he has 
painted his picture, with, it is unnecessary to say, 
appalling results. 

In the book on oil painting in this series by Mr. 
Solomon, R.A., and the book written recently by Mr. 
Harold Speed, we find serious attempts to bring the oil 
painter back to the study of his art as a craftsman. He 
must not be satisfied merely with the knowledge of how 
to express himself in paint, but must also study the 
properties of his materials and their limitations, and so 
modify his methods as to produce not only a beautiful 
picture, but a good job which will stand up to the test 
of time. 

In this book I have tried as far as possible to deal 
rather -^ith methods than materials, onl}^ describing 
materials in so far as information will be of use to the 
artist, endeavouring to A\Tite something useful to the 
artist craftsman, and as far as I can to guide him as 
a craftsman to the proper use of his materials. 



In order to do this I have had as far as possible to 
recover traditional methods, and combine them with 
modern practice. I have also avoided the descrip- 
tion of new and untried materials. The slow chemical 
actions over long periods of time and exposure to light, 
moisture, and oxygen, and the gradual physical changes 
taking place, are so obscure that it is safer to rely on 
materials which have been tested and used throughout 
the centuries than to assume too quickly that new 
materials and new methods will stand the test of time. 

Let the painter in oil study his medium by making 
simple experiments for himself on the amount of oil 
necessary, the effect of darkness on the ground, and 
of the thickness of the layer of oil paint and so on. 
By shutting up his experiments in a box together with 
a jug of water, he will be able in a few months to obtain 
results which will guide him in the choice of his palette 
and in the amount of oil advisable. The way in which 
he should lay on his paint will become more evident, 
and on these lines he should be able to obtain good 
craftsmanship without cramping his style. There is no 
universal method of overcoming the bad properties of 
the oil medium. It is a good servant, but a bad master. 

It is evident that when we look at the whole history 
of painting in oil, the method which has proved most 
successful in avoiding lowering of tone throughout the 
centuries, has been the method in which the oil paint 
is translucent and is painted over a layer of white or 
bright light colour, whether in gesso, tempera, or oil, 
a method which differs widely from the method of using 
oil paint which is most universal to-day, in which solid 
opaque paint is aimed at. The artist, painting in this 
modern method, would be well advised, in my opinion, 
to limit his palette to the more opaque pigments as 
shown in the table given in Chapter IX. By doing this 

Q 241 


and avoiding excess of oil he should be able to avoid 
serious lowering of tone. 

The chapters on tempera, fresco, and wall painting 
I have tried to make useful by combining a description 
of the older methods with an account of modern practice 
from somid practitioners. Doubtless different painters 
have different methods, but in each case the method 
selected is that of a sound craftsman, which should 
therefore prove a useful guide. 

The water-colour painter has three problems — ^the 
good and pure quality of his paper, the permanency of 
his pigments, and the absence of an excess of hygroscopic 
agents like honey and glycerin in his medium. He can 
obtain excellent and pure paper and select a list of 
reliable pigments, and his medium — gum-arabic — has 
also proved reliable under the test of time. 

Unfortunately the demand for moist colours requires 
the loading of the medium with hygroscopic materials 
like glycerin, ^^^lile this has not been proved to be 
injurious, it is at least probable that this practice will 
diminish the life of modern water-colour pictures as 
compared with the older ones that were painted with 
cake colours. It is a matter well worthy of the considera- 
tion of the Water-colour Society, whether a return to 
the old cake colours is not possible. 

In the discussion of fresco and tempera, I have been 
much indebted to modern artists and to the publications 
of the Tempera Society. The test of centuries has proved 
egg to be an excellent and reliable medium, so that the 
question as to how far present-day methods of expression 
can be carried out in tempera is of profound interest to 
the artist. 

In the course of this book I have here and there 
thrown out suggestions to the artists' colourman. On 
the subject of cracking, I have contented myself with 



advice to the artist. The causes of cracking, from the 
point of view of the artists' colourman who grinds pig- 
ments in oil, are only beginning to be understood. The 
properties of the different drying oils, the proportion of 
pigment to oil, the laying of hard coats over soft coats, 
and the size and the grading of the pigment grains, all 
have their influence, and when thoroughly understood 
may result in profoundly modifying the methods of the 
preparation of colours for use. The diminution of the 
yellowing of the oil film by the selection of the most 
suitable oil, and the introduction of resin, is also worthy 
of very careful study. It would be a great gain if, with 
the object of reducing the yellowing of the medium, a 
highly resinous vehicle could be made which at the same 
time is workable under the brush. 

In conclusion, I hope that this book will be a useful 
contribution towards the restoration of the painter 
in oil to the position he once held as a craftsman, 
thoroughly acquainted with the properties of his 
materials and with the right wav to use them. 



Abney, Captain, 90 
Acacia Arabica, 172 
Advocates' Library, Edinburgh, 

43, 97 
iEtius, 18, 133 
Alberti, preparation of wall for 

fresco, 202 
Albumen, egg, 178, 179, 185 
Aleurites cordata, 134 
Alizarin, 97 
Amber, 168 
American Walnut Growers' 

Association, 133 
Ammonia, use of, as solvent, 233 
Amsterdam Museum, method 

of relining, 229, 235, 236 
Angelo, Michael, unfinished 

pictures, 146, 147, 161 
Armenini, Giovanni Battista, 

De Veri Precetti della Pittura, 

Art gallery, rules for, 227 
Athos, Moimt, MS., 37 
Atmosphere, effect on frescoes, 

192, 218, 220, 226 

— effect on pictiu-es, 227, 228 
Aiireoline, 89, 90 

Azurite, 95 

— use of, in fresco, 198, 222 

— condition of, on West- 
minster tombs, 224 

Backing of canvas, 74, 78, 139, 

Balsam, Copaiba, 164 
Barium Chromate, 90 
Baryta water, use in fresco, 209 
Batten, J. D., 73 

Beeswax, 43, 78, 171 

Begue, Le, translation by Merri- 

field, 28, 37 
Beitrdge zur Entivickelungs- 

Geschichte der Maltechnik, 

Berger, 39, 44, 190 
Bending of rays of light, 102, 

Berger, Ernst, Professor, 39, 

44, 190 
Beurs, Willem, early Dutch 

painting methods, 152 
Black pigments, 84 
Bleaching of oil in sunlight, 38, 

39, 41 

— of pictvu'es in north-light, 231 
Blisters, removal of paint, 229 
Blue pigments, 93, 95, 96 
Boiled oil, 130, 131 
Bolognese MS., 44 

Brown, Madox, frescoes, 219 
Buon-Fresco, 17, 191, 193, 195, 

197, 203, 216, 220, 221 
Burns, Robert, 18 
Burnt sienna, 85 
Burnt umber, 85 

Cadmium scarlet, 88 

— yellow, 88 

Callitris quadrivalvis, 167 
Canada balsam, 165 
Canvas, 65, 66 

— backing of, 74, 78, 139, 235 

— priming of, 66, 67, 70, 75, 79 
Cartoons, use of, for fresco, 

Casein, 19, 54, 175 

— use of, in fresco, 219 



Casein, use of, as tempera 
ground, 73 

Cennini, Cennino, 18, 20, 29, 
41, 55, 61, 63, 65, 72, 130, 
147, 148, 173, 179, 181, 183, 
196, 199, 208, 218, 222 

Ceresin, 172 

Cerulean blue, 95 

Charcoal black, 84 

Chemistry oj Paints and Paint- 
ing, Sir A. Church, 80, 210 

Chinese vermilion, 87 

Chlorinated hydrocarbons, use 
of, 233 

Chrome green, 92 

— yellow, 89 

(from lead), 101 

Church, Sir Arthur, 47, 55, 83, 
90, 167, 171, 181, 198, 209, 
215, 218 

Cicinimi, Oleimi, 23 

Clay, etc., for priming, 70 

Cleaning of pictures, the, 227- 

Cleopatra, physician, 3rd cen- 
tury, 42 

Cobalt blue, 94 

— green, 92 

— violet, 96 

— yellow, 89 

Colour, lowering of, 20, 122, 
129, 135, 137, 143, 149, 150, 
156, 192, 220, 227, 231 

— optical properties of , 1 1 1-127 
Colour, Sir A. Church, 127 
Coloiu-s, test of durability of, 

159, 162 

Complementary coloiu-s, 127 

Conegliano, Cima da, un- 
finished picture of, 146 

Copaiba balsam, 164 

Copal, 168 

— varnish as a medium, 159 
Cope, Sir Arthur, 141 
Copperas, 33 

Correggio, picture attributed to, 
examination of, 148 

Dammar, 167, 168 

— varnish, 137 
Dioscorides, 23, 25 
Drying agents, 33 

Dr^-ing of oil films, the, 20, 24, 

40, 156 
Duncan, John, R.S.A., 64, 65, 

181, 183 
Dust, exclusion of, 228 
Dutch, early, methods of 

under-painting, 151 

— pink, 96 

Earth pigments, 85, 86 

for use in fresco, 198 

Eastlake, Sir Charles, 26, 32, 

39, 40, 44, 49, 221, 222 
Egg medium, 17, 18, 177-184, 

186, 188, 190 

optical properties of, 108 

Egyptian blue, 95 

— coffin lid, preparation of, 62 
Eibner, Professor, 42, 43, 76, 

86, 131, 134, 135, 136, 180 
Emerald green, 93 
Emulsions, 185-190 

— early evidence of, 36, 37 

— egg and oil, 43, 48 
EracUus, 23, 28, 49 
Eugenol, 181 

Eyck, the brothers Van, 19, 20, 
21, 34, 40, 42, 44, 46, 49, 
144, 145, 150, 154, 189-190 

Fig-juice, 179 

Flake white, 80, 82, 110, 161 

Florentine pictiure cleaners' 

methods, 234 
Forli, Melozzo da, fresco in 

Vatican, 198 
Formalin, 218-219, 220 
Fresco, 191-226 

— baryta water, use of, in, 209 

— Buon, 17, 191, 195, 203, 216, 

— casein, use of, 218 

— flesh painting in (Vasari), 208 



Fresr-o, Giotto, frescoes of, 208, 
209, 217 

— hatcliing, 209 

— jointing of colours, 205 

— northern tradition, 221, 222 

— oil medium, 29, 225 

— pigments for, 197-199, 207, 

— plaster of Paris, 193-197 

— Pompeiian, 17, 194, 202 

— Roman methods of, 193, 
194, 221 

Fugitive pigments, 100 

Galen, 25 • 

Gallery, rules for a picture, 

Gesso, 34, 61, 63, 65, 73, 142, 

147, 148, 161, 163 
Gilding, 64 

Giotto, frescos of, 208, 209, 217 
Glue, 61, 69 

— gelatine, 64 

— medimii, 19 
Glycerin, 75 
Graccus, Marcus, 42 
Green pigments, 92, 93 
Guevara, preparation of walls 

for fresco, 202 
Gimi-arabic, 17, 172 

Hart, P. Tudor, 182, 190 

Hatching in fresco, 209 

Heaton, Noel, 17, 221 

Hempseed oil, 20, 32 

Hermeneia, the, 41 

Herringham, Lady, transla- 
tions of, 29, 63, 184 

History and Methods of Ancient 
and Modern Painting, Ward, 

Holbein, 47 

— use of gold leaf and yellow 
ochre, 89 

Honey, use of, in water-colotir, 

Hooch, Pieter de, 152 

Hunt,Holman, a'^count of Pre- 
RaphaeUte methods, 71 

— experiments of, 89, 129 

Indian red, 85 

— yellow, 91 
Indigo, 96 
Ivory black, 85 

Keim, improved process of 

stereo-chrome fresco, 215 
Kremnitz white, 80 

Lac, 168 

Lakes, 96, 97, 98 

Lamp black, 84 

Larch (Venice turpentine), 165 

" Last Supioer," Leonardo da 

Vinci, 193 
Latombe, Abraham, 69 
Lead sulphate, 83 

— wliite, 80, 82, 110, 161 
Leighton, Sir Frederick, fres- 
coes, 220 

Lime, preparation of, for fres- 
coes, 17, 195, 196 
Linolenic acid, 129 
Linseed oil, 20, 23, 24, 28, 30, 

34, 128-139, 156 
bleaching of, 38, 41, 130, 

drying of film of, 129, 

137, 142, 143 
oi3tical properties of, 132, 

preparation of, 24, 25, 28, 

30, 34, 128, 129 
Lords, House of, frescoes in, 

215, 218 
Lucca MS., 19, 23, 177 

Maclehose, Louisa, 34, 39, 67 
Madder, 96, 97, 98 
Mahogany panels, 53, 54, 57 
Malacliite, 93 

Manchester, Town Hall, fres- 
coes, 219 



Mander, Van, 150 

Manganese violet, 96 

Mappae Clavicula, 23 

Marble for fresco, 196 

Mars yellow, 86 

Mastic varnish, 163, 167, 228- 

Massicot, 91 
Materials for a History of Oil 

Painting, Eastlake, 26, 39 
Mayerne, De, 69, 70 
McEwan Hall, Edinburgh, 

frescoes in, 219 
Merrifield, Mrs., 28, 37, 39, 191 
Merritt, Lee, Mrs., frescoes at 

Wonersh, 215 
Morrell, Dr., 136 
Mount Athos MS., 37 
Mural painting, 17, 29, 191-226 

Naples yellow, 91 

Oil, darkening of, 39, 40, 124, 
129, 137, 142, 146, 241 

— drying of a film of, 20, 24, 
40, 156, 157 

— medium, early methods and 
discovery, 19, 20, 22-48, 144, 
148, 151 

— optical properties of, 140- 

— poppy, 20, 76, 134, 151, 152, 

— preparation of, 24, 25, 28, 
32, 34, 40, 128 

— priming for canvas, 68, 69, 
70, 71, 73, 156 

— spike of lavender 

— Stand, 41, 42, 77, 137, 138, 

— Tung, 134 

use in fresco of, 218, 220 

— varnishes, 170 

— walnut, 20, 133 
Oleo d'Abezzo, 45 
Oleum Cicinimi, 23 
Opacity of pigments, 119, 120 

Optical principles, 102-127 
Orange cadmimii, 88-89 

— chrome, 89 
Orpiment, 90 

Oswald, Professor, 219, 229 
Oxide of chromium, 92 

Palette for fresco painting, 215 
Panels, 53, 54, 58, 60 

— priming of, 59-65 
Paper, 51, 52 
Paraffin wax, 172 
Parchment size, 174-175 
Parry, Gambler, mediimi for 

fresco, 219, 220 
Permeation of moisture, 45, 55, 

Pentimento, 109 
Petrie, Professor Flinders, 19 
Petroleum, 171 
Pigments, 80-101 

— early, 46, 47 

— effect of changes in linseed 
oil on (table), 124 

— fresco, 197, 199, 207, 212, 

— permanent, 98, 99, 161 
Pigments and Mediums of the 

Old Masters, Professor A. P. 

Pink, Dutch, 96 
Pinus balsatnea, 165 
Pinus maritima, 166 
Plaster (fresco), 193, 195, 197, 

201, 202, 204 
Pliny, 18, 23, 42, 80, 93, 177 
Ply-wood, 54, 57, 58 
Pom peiian frescoes, 17, 194 
Poplar, white, for panels, 53 
Poi^py oil, 20, 76, 134, 151, 152, 

Potash, sihcate of, for fresco 

painting, 215 
Practice of Oil Painting, S. J. 

Solomon, R.A., 151, 152 
Pre-Raphaelite teclinique, 71, 

72, 240 



Preservation of pictures, 227- 

Priming, 27, 59-65 
Prussian blue, 93 

Raphael, 52 

Raw sienna, 85 

Raw umber, 85 

Red ochre, 85 

Reflection of Ught, 102-103 

Refraction of light, 102-127 

Refractive index (formula), 103 

ReUning of pictures, 229, 235- 

Resins, 166, 167 
Restrainers, 234 
Revnolds, Sir Joshua, 47, 90 
Rich, A. W., 52 
Ripolene enamel, 59 
Rokeby Venus, the, 150 
Rubens, letters of, 130 

— on effects of sunlight, 38 

— portrait of Susanne de Tour- 
ment, 153 

RusseU, Professor, 90, 97 

St. Marj-'s Cathedral, Edin- 
burgh, frescoes, 225-226 

St. Stephen's Chapel, decora- 
tion of, 221 

Sandalwood oil, 237 

Sandarac, 167, 221 

Sante Croce, Florence, frescoes, 

Sap green, 93 

Saponin, 231 

Scahng, detection of, 228 

Seasoning of timber, 55 

Sienna, biurned and raw, 85, 86 

Size, 17, 34, 51, 174 

— optical properties of, 108 

— painting, 223, 224 
Sleen, Van der, 100 
Smalt, 96 

— in fresco, 198 

Solid high lights, tradition of, 
152, 153 

Solomon, Solomon J., 151, 152, 

Solvents, 232, 233 
Spectriun, the, 111-127 
Speed, Harold, 240 
Spike, oil of, 166 
Spirits of turpentine, 165 
Spirit varnishes, 170 
Stand oil, 41, 42, 77, 137, 138, 

Stereo-chrome process of fresco, 

Strasburg MS., 32, 41, 164, 223, 

Simlight, effect on oil of, 38, 39, 


Talbot, L. Agnes, parchment 

size, 174 
Tempera painting, 177-184 

— restoration of pictures in, 

— Societv, Transactions of, 73, 
174, 184, 190 

Temperature, importance of, 

Terre Verte, 85 

in fresco, 198 

Theopliilus, 20, 26, 48, 133 
Thickened oils, 41, 42 
Tliickness of panels, effect on 

warping, 58 
Titanium wliite, 83 
Traquair, 220, 222, 225, 226 
Transparencv of pigments, 119, 

120, 161, 163 

table, 124 

Tristram, E. W., examination 

of Westminster tombs, 223, 

Timg oil, 134, 218, 220 
Turpentine, spirits of, 165 

— Strasbiu-g, 164 

— Venice, 36, 45 

Ultramarine, 94 

— sickness, 100 



Umber, 85 
Underpainting, in egg, 21 

— in grey, 151, 152 

— flesh, Michael Angelo, 147, 
161, 184 

— flesh, Velasquez, 150 

Vandyke brown, 86 
Varnish, 33, 35, 36, 164, 169, 
170, 176 

— addition of, to oil medium, 
44, 45 

— removal of, 231 

— Tempera (Cennino Cennini), 

Varnishes and their Components, 

Dr. Morrell, 171 
Varnishing, the, of a picture, 

Vasari, 20, 21, 34, 36, 40, 44, 

67, 68, 133, 177, 186, 202 

— on Van Eyck, 49 

— Life oj Alessio Baldovinetti, 

Velasquez, Rokeby Venus, 150 
Venetian red, 85 
Venice turpentine, 36, 45, 164 
Verdigris, 93 
Vermilion, 87 
Vernition, 25 
Verte emeraude, 92 
Vinci, Leonardo da, " Last 
Supper," 192 

— use of massicot, 
Violets, the, 96 
Viridian, 92 

Vitruvius, preparation of wall 
for fresco, 193, 202 

Walls, preparation of, for 

fresco, 193, 195, 197, 201, 

204, 205 
Wall painting, 191-226 
Walnut oil, 20, 133 
Ward, James, History and 

Methods of Ancient and 

Modern Painting, 19 
Warping of panels, 54, 58 
Water colours, 173, 175, 242 
Wax, beeswax, 171 
-I— ceresin, 172 

— medium, 18, 19 

— medium for fresco, 215-219 
Westminster, the painted 

tombs at, 223 
White lead, 80, 82, 110, 161 

— hght, behaviour of, 102-110 

— pigments, 80-84 

Whites, transparency of the, 

Wood-beetle, 228, 237 
Wood panels, 53, 54, 58, 60 

Yellow pigments, 88 

— chrome, 89 

— cobalt, 89 

— Indian, 91 

— Naples, 91 

— ochre, 85 

— oxide of lead, 91 
Yellowing of oil, the, 122, 124, 

129, 135, 136, 137, 143, 145, 
149, 150, 156-159 

Zinc chromate, 90 

— white, optical properties of, 
109, 110 


Seeley, Service & C 

^ Founded I7QC 




"The admirable New Art Library."— Connoisseur. 


By Professor A. P. Laurie, m.a. ,, 
K. U.S. E., HON. U.S.A., Profcssor of Chemistry 
to the Royal Academy. With 50 illus. 21s. nett 

LANDSCAPE. Hy Adrian Stokes, k.a. L'S^. nett. 
ETCHING. I'V K. S. Lumsdex. 2o3 illus. 21s. nett. 
DRAWING. By Harold Speed. 96 illus. los. 6d. nett. 
OIL PAINTING. S. J. Solomon, r.a. So illus. 10s. 6d. nett. 
ANATOMY. Sir Alfred D. Fripp. 159 illus. 15s. nett. 
MODELLING &- SCULPTURE. A. Toft, ng illus. 15s. nett. 
WATER COLOUR. A.W.Rich. 60 illus. los. 6d. nett. 
PERSPECTIVE. R. Vicat Cole. 472 illus. 15s. nett. 
ANATOMY OF TREES. R. V. Cole. 500 illus. 15s. nett. 

Hi miration Jrom, Speed's " Drawing. 



16. WIRELESS OF TO-DAY. By C. R. Gibson 
& W. B. Cole. 6s. nett each Vol. 

Some of the Volumes. 

A Marine Reptile. 
From "Geology of To-day. 


C. R. Gibson, f.r.s.e. 


Cecil G. Dolmage, m.a., 
d.c.l., ll.d., f.r.a.s. 


DAY. C. R. Gibson, F.R.S.E. 


G. F. Scott Elliot, m.a. 


T. W. Cor bin. 


Prof. H. Chapman Jones. 


C. W. Domville-Fife. 


J. \V. Gregory, f.r.s. 


Maj. C. C. Turner. 

C. A. E ALAND, M.A. 

P. G. Bull, m.a. 


Cloth, 3s. 6d. nett. Leather, 5s. nett. 
" Dainty and attractive little books."— Daily Post. 

THINGS SEEN IN ROME. A. G. Mackinxon, m.a. 
THINGS SEEN IN NORWAY. S. C. Hammer, m.a.(oslo). 


!:nin j 


1. JAPAN. Ci-ivB 
•2. CHINA. J. R. Ci! 

3. EGYPT. E. L. Bt 

4. HOLLAND. C. E. Roche. 

5. SPAIN. G. Hartley. 

6. NORTH INDIA T.L.Pennell 

7. VENICE. Lonsdale Ragg. 
0. PALESTINE. A. G. Freer. 


10. OXFORD. N. J. Davidson. 

11. SWEDEN. Barnes Stevemi. 

12. LONDON. A. H. Blake. 

13. FLORENCE. E. Geibrson. 

14. ITALIAN LAKES. L. Ragg. 

15. RIVIERA. L. Richardson. 


Clive Holland. 



A. Goodrich-Freer, 

18. EDINBURGH. E. Grierson. 


C. ^\■. DoMVILLE-FlFE. 


\V. T. Palmer. 

21. PARIS. Clive Holland. 




By Commander E. Hamilton Currey, r.k. 

With Coloured Illustrations. Extra Crown 8vo. 5s. nett each. 

Eacli of the volumes in the Ian Hardy Series is a complete 
storj' by itself, but each describes another stage of the 
adventurous life of the delightful young rascal who is their 



"An ideal gift book for a boy, for it is utterly 
devoid of the milksop piffle in which such stories 
usually indulge. A whole budget of mirth and adven- 
ture." — Manchester Courier. 


"Combines as much adventure as the most spirited 

of young readers could possibly wish for, with a 

faithful picture of what a middy's life may be when 

he is lively and as plucky as tan. '—Glasgow Heraid. 


"It is no exaggeration to say that Commander 
Currey bears worthily the mantle of Kingston and 

Marryat. "—Manchester Courier. 


"Commander Currey is becoming a serious rival to 

Kingston."— Yorkshire Observer. Specimen Binding. 


B()isii.\GON-. Coloured Illustrations. 6s. nett. 
"A dashing youth who has a good time, but emerges from his training with a good 

character." — Manchester Courier. 


Lavishly Illustrated. Ex. Crown iSvo. 6s. nett each volume. 

" Splendid volumes." — The Outlooli. 

"Gift books whose value it would be difficult to over-estimate. ' — The Standard. 

'This series has now won a considerable and well-deserved reputation." — The Guardian. 




1 . W. COKBIN. 


-A. \\ ILII.-.MS, F.R.G.S. 


.•\. Williams. 


M. LOCOMOTION. A. Williams. 

C.K. Gibson, F.R.s.E. 
MOD. MECHANISM. A. Williams 
INSECT LIFE. E. Selous. 
MINING. A. Williams, b.a. 
BIRD LIFE. loHN Lea, m.a. 
CHEMISTRY. Prof. Philip, 

_ -.-,,« ^ MANUFACTURE. C. R. Gibson. 

%k /jp>- EARLY BRITISH LIFE. By Prof. 

-; '•^ ^ Sc"Ti Elliot, .m.a., 

GEOLOGY. E. S. Gre«-, m.a. 

John LtA, m.a. 

Qr Krii> 



.A. WiLLIA.MS, F.R.G.S. 


Dr. Lambert, m.a. 
MIGHTY DEEP. A. Giberne. 

G. F. Scott. 
SAVAGE LIFE. Prof. S. Elliot. 
SIEGES. E. GiLLiAT, m.a. 
THE SHIP. KebleChatterton 
ASTRONOMY. H. Macpherson 

C. R. GiBSO.V, F.R-S.E. 

PIRACY. Lieut. K. Chatterton 


COMMERCE. H. O. Nkwland. 

C. A. E ALAND, M.A. 

RAILWAYS. T. W. Corbin. 
COAL. C. R. Gibson, f.r.s.e. 

K. K. Chatterton, b.a. 
T. W. Cokbin. 



Popular Science for Young Children of 8 to 14. 
Each 5s. nett. 
" Among writers on science for boys, easily the most 
skilful is Mr. Charles Gibson."— The Nation. 

NKW volump: 

C. R. GlHSON, F. U.S.E. 

PHOTOGRAPHY 3-^ ITS MYSTERIES. C. R. f.ipsfix, f.k.s.k. 
THE GREAT BALL ON WHICH WE LIVE. C. R. Gibson, f.r.s.e. 
"An admirable introduction to about half a dozen 
of the most recondite sciences without going beyond 
the comprehension of any intelligent chWd."— Scotsman. 
" Reads like a fairy tale, yet every word is true. " 

Dundee Courier. 

C. R. Gibson, f.r.s.e. Specimen Binding. 


Particularly handsome Gift Books for Young People. Illustrated. 3s. nett each. 


B.A., F.R.G.g. 


liams, F.R.G.S. 

liams, B.A. 

INGENUITY. H Coupin,, and John Lea, m.a. 

PLANT WORLD. Prof. (,. 

K. Scott Klliot. 


Williams, b.a. 

LIFE. John Lea, m.a. 


Prof. J. C. Philip, f.r.s. 

TRICITY. C. R. Gibson. 


Williams, b.a. 

tor Macpherson, m.a. 

AGE LIFE. Prof, .--cott 

Elliot, m.a., 


C. R. Gibson, f.r.s.e. 


With many Illustrations in Colours. 5s. nett each. 


1. FAMOUS PIRATES. Lieut. -Commdr. Keblk 

Cmatterton, B.A.(Oxon). 


3. INDIAN MUTINY. Edward Gilliat, m.a. 

4. DARK FORESTS. H. W. G. Hyrst. 

NoR.MAN J. Davidson, B.A.(Oxon.) 


5. GREAT PATHFINDERS. E. Sanderson, m.a. 
0. GREAT MOUNTAINEERS. Richd. Stead, b.a. 
7. POLAR EXPLORERS. G. Firth Scott. 



With Jackets in 3 colours & Illustrations. 6s. n. each. 



Barbrooke Grubb, Bishop Bompas, &c. R. M. A. Ibbotson. 
"Of all books on Missions we have read this is the most fascinating." — Dundee Courier. 


Adven'ti'res on the I Adventi'res on the 

1. HIGH MOUNTAINS. R. Stead, B.A.,F.R.H.s. 6. HIGH SEAS. R. Stead, b.a. 

2. GREAT FORESTS. H. W. G. Hyrst. 7. ARCTIC REGIONS. H. W. G. Hyrst. 

3. GREAT DESERTS. H. VV. G. Hvkst. S. RED INDIANS. H. W. G. Hyrst. 

4. GREAT RIVERS. R. Stead, b.a., f.r.h.s. 9. TRAPPERS «5 HUNTERS. K. Young, 

5. WILD BEASTS. H. W. G. Hyrst. 1 10. SOUTHERN SEAS. R. Stead, b.a. 


With Covers in 4 Colours. Illustrated. Price Is. nett. 



"With attractive coloured covers, well written, and at the wonderfully low price of 
one shilling. Will bring delight to the boys and girls who get them."— Missionary Herald. 


MOFFAT OF AFRICA. N. J. Davidson. 

ARNOT OF AFRICA. Nigel B. M. Grahame. 






JUDSON of BURMA. Nigel B. M. Grahame, b.a. 


Handsome Gift Books for Young People. Many Illustrations. 4s. 6d. nett each. 

17. THE COUNT OF THE SAXON SHORE. Prof. A. J. Church. 




3. THE FALL OF ATHENS. Prof. Church, m.a 

4. THE KING'S REEVE. Kdward Gii.liat, m.a. 

5. CABIN ON THE BEACH. M. E. Winchester. 


7. GRIMM'S FAIRY TALES. New Translation. 

8. THE WOLF'S HEAD. Edward Gii.liat. m.a. 

10. CAEDWALLA. Frank Cowper. 
1-2 ROBINSON CRUSOE. Daniel Defoe. 


14. CRANFORD. Mrs. Gaskell. 

15. PATRIOT AND HERO. Prof. Church. 

16. MINISTERING CHILDREN. Charlesworth. 



Each Volume fully Illustrated. Ex. Crown 8vo. 6s. nett. 



Edward Gii-mat, m.a., sometime 
Master at Harrow School. 

With Full-page Illustrations. 

"A capital book. . . . Full of stories of heroic 
deeds, intrepidity, and determination performed 
in the face of fearful odds." — The Field. 

Heroes of 

1. MISSIONABY ENTERPRISE. Claud Field, m.a. 

2. PIONEERING. Edgar Sanderson, m.a. 

3. MISSIONARY ADVENTURE. Canon Dawson, m.a 

4. MODERN CRUSADES. E. Gilliat, m.a. 

5. MODERN INDIA. E. Gilliat, m.a. 

6. THE ELIZABETHAN AGE. E. Gilliat, m.a. 

7. MODERN AFRICA. E. Gilliat, m.a. 

8. THE SCIENTIFIC WORLD. C. R. Gibson, f.r.s e. 

The Storming or Jhansi. 

From "Heroes of the Indian Afuliny." 


With many Coloured Illustrations, .^is. nett each. 

7. CRANFORD. Mrs. Gaskeli.. 

A KNIGHT ERRANT. N. J. Davidson, b.a. 

THE FAERY QUEEN. Prof. A. J. Church 


Books for Boys Sf Girls. Illustrated. 2s. 6d. nett each. 






THE PIRATE. Capt. Marryat. 


LIONHEAHTED. Canon E. C. Dawson. 

TO THE LIONS. Prof. A. J. Church. 


MISSIONARY HEROES IN ASIA. J. G. Lambert, m.a., d.d. 

LITTLE WOMEN. L. M. Alcott. 

GOOD WIVES. L. M. Alcott. 


GRIMM'S FAIRY TALES. A New Translation. 

WHAT KATY DID AT HOME. Susan Coolidge. 


BoxE.s from Home. 

'■'rom •'What K'aty Did." 



With many Illustrations and a Frontispiece in Colours, 3s. 6d. nett each. 

11. BARBROOKE GRUBB, PATHFINDER. N. J. Davidsov, B.A.(Oxon). 

"One of the best known of missionary pioneers to South America . . . Will appeal to 
that spirit of adventure which is in the heart of every boy or girl." — Record. 


Vol. 3. 

Vol. 4. 

Vol. 1. 

Vol. 2. 


1. A HERO OF THE AFGHAN FRONTIER. The Life of Dr. T. L. Pennkll, of 

Bannu. \. I\L Pennell, m.b. , B.s.(Lond.), 

2. MISSIONARY CRUSADERS. Rev. Claud Field, m.a. 

3. JUDSON. THE HERO OF BURMA. Jesse Page, f.r.g.s. 


Life amongst Red Indians e~ Esquimo. Rev. H. A. Cody, m.a. 

5. MISSIONARY KNIGHTS OF THE CROSS. Rev. J. C. Lambert, m.a., d.d. 



8. BY ESKIMO DOG-SLED & KYAK. By Dr. S. K. Hutton, m.b., cm. 

10. ARNOT, A KNIGHT OF AFRIC.'V. Rev. E. Baker. 

Binding oj the Russell Stri, 


A Series of handsome Gift Books for Boys &= Girls. 
With Coloured and other Illustrations. 3s. 6d. nett each. 



A. Williams, B.A.(Oxon). 


Stirring narratives of bravery and peril. H. W. G. Hvrst. 


E. Gilliat, m.a., sometime Master at Harrow. 





C. R. Gibson, f.r.s.k. 




" Mr. Gibson has fairly made his mark as a populariser of scientific luiowledge."— G^iuirdiuTt. 
" Mr. Gibson has a fine gift of e:E.^oaition."— Birmingham I'ost. 

In the Scienck of To-Day Sekies. Illustrated. 6s. nett each. 
SCIENTIFIC IDEAS OF TO-DAY. A Popular Account of the Nature 

of Matter, Electrifity, Light, Heat, ©"c. &c. 
WIRELESS OF TO-DAY. By Charles R. Gibson, f.r.s.e., & W. B. Cole. 

In the Ro.^iance Library. Illustrated. 6s. nett each. 

WHAT IS ELECTRICITY ? Long 8vo. With 8 Illustrations. 6s. nett. 



' CR OIB'iClN 





By CHARLES R. GIBSON, f.r.s.e. 
Extra Crown 8vo. Fully illustrated. 5s. nett each. 


VOL. I. 


makes scientific matters clear and Inter- 
esting. This book is ALL THAT THE 

The Lady. 

TATION. Every experiment is so clearly 
explained and skilfully handled that the 
lesson is learnt at once without difficulty. 
He has such an engaging way of Imparting 
Ills knowledge that he is entertainingly 
educative. An admirable gift book, 
packed with illustrations showing how the 
experiments can be carried cut without 
much trouble or expense " 

Yorkshire Observer. 


"The Headmaster of Eton (Dr. the Hon. E. Lyttelton) advised his hearers, in a recent 
speech at the Royal Albert Institute, to read Professor A. J. Church's ' Stories from Homer,' 
some of which, he said, he had read to Eton hoys after a hard school day, and at an age 
when they were not in the least desirous of learning, but were anxious to go to tea. The 
stories were so brilliantly told, however, that those young Etonians were entranced by them, 
and they actually begged of him to go on, being quite prepared to sacrifice their tea-time." 
Profusely illustrated. Extra Crown 8vo, 5s. nett each. 

The Children's ^Eneid 

The Children's Iliad 

The Children's Odyssey 

The Faery Queen &=her Knights 

The Crusaders 

Greek Story and Song 

btories from Homer 

Stones from Virgil 
The Crown of Pine 
Stories of the East from 

Story of the Persian War 
Stories from Livy [Cicero 

Roman Life in the Days of 

Count of Saxon Shore 

The Hammer 

Story of the Iliad 

Story of the Odyssey 

Heroes of Chivalry & Romance 

Helmet and Spear 

Stories of Charlemagne 



A liUle Book lor Wayfaring Men. In Prose and Verse. Fcap. 8vo. cloth. 38. 6d. nett ; thin 
paper. 5b. nett ; bufl leather yapp. In a box. 68. nett. 

"The very flower of a cultivated man's reading."— Coitn(r2/ Life. 

EDINBURGH. By R L. Stevexson. 

The lollowlng are the various Editions : 1. Crown 4to. With Illustrations in colour by James 

Heron. I0s.6d.nett. 2. Library Edition, buckram. Fully Illustrated. Crown 8vo. 7a. 6d. n. 

3. Pott 4to. With Coloured Illustrations. 68. nett. 4. Fcap. 8vo. With Frontispiece, cloth. 

28. 6d. nett, leather. 48. 6d. nett. 6. People's Edition. Illustrated. Demy 8vo. Is. nett ; cloth. 

28. nett. 

Only 355 
258. nett. 

OXFORD. By Andrew Lang. 

The following are the various Editions. With many Illustrations by A. Brunet Debaikbs, 
A. Toussaint. and R. Kent Thomas. 

1. Crown 4to. Printed by Messrs. T. & A. Conotabi-e of Edinburgh. 128. 6d. nett 

2. Edition de Luxe, on unbleached Arnold hand-made paper, each copy numbered. 

have been printed, of which 350 are for sale. 

3. Pott 4to. With Coloured Illustrations. 79. 6d. nett. 

4. Ex. Crown 8vo. With Coloured Frontispiece and Fifty other Illustrations. 78. 6d, 
6. Crown 8vo. Illustrated. 48. nett. 


Fully Illustrated. Each Volume, 4s. nett. 

By C. A. Ealand, m.a. 



By T. W. CoRBiN. 

2. AVIATION. Maj.C. C. Turner. 

5. GEOLOGY. By E. S. Grew, m.a. 


By T. W. CoRBiN. 

5. PHOTOGRAPHY. C.R.Gibson. 


6. THE SHIP. By Lieut-Corn. E. 

KebleChatterton, r.n.v.r. 


By T. W. CoRBiN. 


By Sidney Wright. 

9. RAILWAYS. By A. Williams. 


Each volume consists of over 700 pages, cloth, 6s. nett each. Specimen Binding. Thin Paper Edition (measuring 6|x4J ins., and only | of 

an inch thiols), price "s. 6d. nett each. They also contain practical Introductions by 
Prof. Cairns James, Professor of Elocution at the Royal College of Music. 


Selected from the works of Rudyard Kipling, R. L. Stevenson, Conan Doyle, 

Thomas Hardy, Austin Dobson, Christina Rossetti, Maurice Hewlett, A. W. 

Pinero, Sydney Grundy, &c., &c. 

" An admirable collection of pieces both in pro.'se an'l verse." — Spectator. 

" Offers an unusual wealth from authors of to-day." — The Athenceum. 

" Far superior to anytliing we have yet seen." — Western Press. 


From Anstey, Barrie, Major Drury, J. K. Jerome, Barry Pain, A. W. Pinero, 

Owen Seaman, G. Bernard Shaw, &c. , &c. 

"A most comprehensive 6^ well-chosen collection of some hundreds of pieces. A small 

encycloppedia of Euglisli humour." — The Spectator. 

" Unquestionably the best collection of modem humorous pieces for recitation which 

has yet been issued." — 'I'he Dundee Adcertiser. 


The Things Seen Series 

Each volume fully illustratet! and bound in cloth. 3s. 6d. nett ; i;i leather, 3s. nett. 

"A successful series by capable writers."— 7"/»e Time!. 

"Beautifully Illustrated with photographs of characteristic scenes and people." 

The Daily Telegraph. 

"As each new volume of 'Things Seen' comes out it serves to show how admirably 
the whole series has been planned ^ how capably the work has been carried out. 
The little books have a character 6- quality of their OMTn."— British Weekly, Dec. 10, '25. 


By A. Goodkich-Freer 

"Brightly written from full and intimate know- 
ledge of the subject." — Scors>'ia?t, 


By N. J. Davidson, B.A.(Oxon) 

" An excellent account of the University town and 
the life of the Undergraduate-" — jloerdetn youriial. 


By E. L. Butcher 

" Mrs. Butcher is thoroughly conversant with her 
subject . . . excellently written."— o7i)*^. 


By C. Gasquoine Hartley 
'This admirable little volume."— 0«/u«*<r. 

Canon & Mrs. Lonsdale Ragg, B.D. 
"By authors who are both of them cultured 
students of Italian life, art, and literature." — Titnes. 


By T. L. Pe.nnell, M.D., B.Sc 

"A delightful account of the sights of Upper India 
by one who has special knowledge-" — Bham Post. 


Clive Holland 

"An attractive vol- 
ume; the photographs 
with which it is illus- 
trated are admirable." 
Manchester Guard. 


W. B. Stbveni 

"A charming little 
volume." — Ch.of En^' 


J. R. Chittv 

" By a writer who 
adds grace and style 
to entire familiarity 
with the country ^ 
peop\&." —£" /Y m Post. 


E. Grierso.s 
"The best of a 
wonderful series." 

Duyidte Courier. 


By C. W. DoMVIl.LE-FlFK 
" .\ Tery complete book on Winter Sports." j 

Western Daily Ne^LS. \ 


By A. H. Blake, M..\.(0.xon) | 

" This charming little book full of excellent illus- | 
\r3.x\oTiS." —Aberdee'i ycumal. 

French &= Italian 

By Capt. Leslie Richardson i 

" Pleasantly written &• with a profusion of ad- 
mirable illustrations." — Sp(c:ator. 

By C. E. RocHB 

"A charming addition to the series . . . eminently 
readable." — Moniins; Post. 


By Mrs. Lonsdale Ragg 

"A delightful addition to the charming Things 
Seen Series."— //Vjterji Morning Anes. 


By Clive Holland 

" Quite the most attractive volume in an admirable 
■ Series."— SrjUsh Ifeeily. 


By E. Grierson 
" Eminently interesting." — Bristol Times. 


i By .'\. Goodrich-Frekr, F.R.S.G.S. 

" Most charming. As readable as it is informative." 
I Aberdeen Press. 


" Gives the spirit, the romance] £-the beauty of the' ^^>' 9"^^ Holland. 

lake country as well as its topography." Illustrated some really distinguished 

Li-.erfcol Ccurier. photography." — Sheffield Daily Telegraph. 

By S. C. Hammer, M.A. 

Profusely Illustrated. 



With Many Illustrations. 

Seeley, Service <Sf Co. Ltd., 196 Shaftesbury Avenue, W.C. 2 





Professor of English Literature at Cambridge University 

Clothy price, nett, 55. Thin paper edition in leather, 7s. 6d. ncit ; huffed leather, 

yapp, in a box, price, nett, 7s. 6d. 

"Prof. Quiller-Couch is the prince of anthologists." 

The Glasgow Evening News. 

"A. little book of grave and beautiful thoughts. It would be difficult to better 

the selections." — I'he Guardian. 

" The poems and prose passages are chosen — as might be safely foretold — 

with taste aud discrimination, and the volume will be found a heartening 

companion." — The Tribune. 

"The very flower of a cultivated man's reading." — Country Life. 

"Prof. Qiuller-Coucli's anthologies are the best of their kind in modern English 

literature." — The Morning Post. 







By prof, cairns JAMES 

Professor of Elocution at the Royal College of Music and the Guildhall School of Muale 

Extra crown %vo, over 700 pages, cloth, nett, 6s. ; also a thin paper pocket edition, 

with coloured edges, nett, 7s. 6d. 

" An admirable collection of pieces, both in prose and verse." — Spectator. 

" Far superior to anything we have yet seen." — Western Press. 

" A more admirable book of its kind could not well be desired." 

Liverpool Courier. 





G. B. SHAW, (Sfc. &G. 


By prof, cairns JAMES 

Extra crown 8t>o, over 700 pages, cloth, nett, 65. ; also a thin paper pocket 

edition, with coloured edges, nett, 7s, Qd. 

"Unquestionably the best collection of modern humorous pieces for recitationa 

which has yet been issued." — The Dundee Advertiser. 

" Packed with things that arc fresh and unhackneyed." — Bookman. 

"An excellent selection, three-fifths of them being taken from the work of the 

best modern writers." — The IVorld. 

" A most comprehensive and well-chosen collection of some hundreds of pieces — 

& most catholic array of all that is good in English literature, and a small 

encyclopaedia of English humour." — 2'he Spectator. 




Hon. D. Ml'9. Oxford, Durham & Wales, 
A Member of Council, Royal College of Music, 


With an Essay on Music &f Musical Criticism, & Four Portraits. 
Eleventh Edition. 5s, nettcach. Also a Library Edition, 8s. 6(?. nett. 
"One more bit of advice is necessary. READ NOAV, IF YOU HAVE NOT 
the main characteristics of beauty in music are clearly defined. Nothing could 
assist you better in the task of ' arranging your emotions ' and obtaining a 
glimpse of the reasons experienced music-lovers have for describing works as 
beautiful or the reverse." — Extract from "Musical Taste & How to Form It," by 

M. D. Calvocoressi. 

"The first attempt towards laying the foundations of a specific method in 
musical criticism is Hadow's ("Studies in Modern Music," London: Seeley, 
Service & Co., First Series), who disengages the four main principles — vitality, 
labour, proportion and fitness — from the existence of which, estimates of musical 
works can be arrived at." — Extract from " Dictionary of Modern Music & 



With an Essay on Musical Form & Four Portraits. 

Eleventh Edition. 5s. nett. Also Library Ediiion,8s. 6d. nett, 

" The development of form is described with MANY BRILLIANT TOUCHES 

and with complete grasp of the subject, and the book, which will probably be 

considered to be EVEN BETTER THAN THE FORMER WORK, is most 

heartily to be recommended to all who wish to attain the highest kind of 

enjoyment of the best music." — Times, 




M.A.(Cantab.), B.Sc.(Edin.), F.R.S.E., F.L.S., F.R.G.S., 
Author of "A Naturalist in Mid-Africa, Qfc, &c., &c. 

Demy 8vo, With 70 Illustrations, 10s. %d. nett, 
" In addition to a marvellously wide range of information, the author is 
" Professor Scott Elliott has contrived to condense all the recent discoveries 

which bear upon the problem of prehistoric man." — Outlook. 




"We could pile up adjectives to express our appreciation of this delightful 

work, and of the broad-minded way in which this great antliropologist tells us of 

the conclusions he has come to." — Fishing Gazette. 



fVith many Illustrations. Extra Crortn Zpo, 6s. nett each. 


WIRELESS OF TO-DAY. By Charles R. Gibson, F.R.S.E., c 

W. B. Cole, With many Illustration! 
" Admirable for the amateur.' —B'ham. Gazette. " Provides all that is necessary." 

Glasgow Herala 

CHEMISTRY OF TO-DAY. A Popular Introduction in Non- 
technical Language. With 150 Illustrations. P. G. Bull, M. A. (Oxen). 
" Combines in a delightful degree lucidity with the strictest adherence to scientific 
accuracy." — Building News. 


Author of "Insects and Man," (Sfc. With 26 Illustrations. 
"A wholly fascinating natural history \>oa)t.." —Outlook. 


Submarine Menace was met fif vanquished. C.W. Domvillk-Fifi. 53 Illustration*. 
"A very striking book, revelation follows revelation, and magnificent stories oi 
fighting &r heroism at sea come practically on every page. One of the few nrar books 
which will survive the next 10 years." — Liverpool Courier. 

AIRCRAFT OF TO-DAY. A Popular Account of the Conquest 

of the Air. By Maj, Charles C. Turner, R.A.F. 62 Illustrations. 
" Maj. Turner is well known as an authority on aeronautics. Of real value." 

A berdeen Journal, 

GEOLOGY OF TO-DAY. A Popular Introduction in Simple 

Language. By J. W. Gregory, F.R.S., D.Sc, Professor of Geology at the 
University of Glasgow. With 55 lUustrationi. 
"An ideal introduction to a fascinating science. The romance and reality of the 
earth most brilliantly and soundly presented.'' — Globe. 


DoMviLLE-Firi, Author of " Submarines of the World's Navies," Sff. 

BOTANY OF TO-DAY. A Popular Account of the Evolution of 

Modern Botany. By Prof. G. F. Scott-Elliot, M.A., B.Sc, F.L.S. 
"This most entertaining and instructive book. It is the fruit of wide reading and 
much patient industry."— G/*^<. 

SCIENTIFIC IDEAS OF TO-DAY. A Popular Account, in 

Non-technical Language, of the Nature of Matter, Electricity, Light, Heat. 

Electrons, &€. &c. By C. R. Gibson, F.R.S.E. 

"As a knowledgeable writer, gifted with the power of imparting what he knows in a 

manner intelligible to all, Mr. C. R. Gibson has established a well-deserved reputation. ' 


ASTRONOMY OF TO-DAY. By Cecil G. Dolmage, LL.D., 

F. R. a. S. 46 Illustrations. 
"A lucid exposition much helped by abundant illustrations." — The Times. 

ELECTRICITY OF TO-DAY. Its Work and Mysteries Ex- 
plained. By Charles R. Gibson, F.R.S.E. Extra Crown 8vo. 
"One of the best examples of popular scientific exposition that we remember seeing.'' 

The Tribune. 

ENGINEERING OF TO-DAY. T. W. Corbin. 39 Illustrations. 

" Most attractive and \asiraci\v^."— Record. 


"In knowledge and clearness of exposition it is far better than most works of a 
similar character and aim." — Academy, 

PHOTOGRAPHY OF TO-DAY. A Popular Account of the 

Origin, Progress, and Latest Discoveries. By H. Chapman Jones, F. I.C, F.C. S. 
"An admirable statement of the development of photography from its very beginning 
to the present tima,"— Journal o/ Photogra/ihy. 



PLATE Facing 

45. The Governors of St. Elizabeth's Hospital ' 

(another portion partially cleaned) . . .232 

46. A Florentine Picture Cleaner's Method . . 234 

47. Micro-Photograph of a Turner in the Tate 

Gallery ....... 236 

48. "The Coming of St. Columba." By William 

M'Taggart, R.S.A 238 



1. Apparatus for Testing the Warping of Panels 56 

2. Ray of Light Passing through a Glass Plate 102 

3. Ray of Light Passing through Transparent 

LlQXHD ....... 104 

4. Ray of Light Passing through two Liquids . 105 

5. Ray of Light Passing through Liquid and Glass 106 

6. Glass Prism Dividing Light . . . .111 

7. Effect of Two Glass Prisms on a Ray of Light 112 

8. "Closing up" the Spectrum with Mirrors . 112 

9. Mixture of Blue and Yellow Producinq Green 115 

10. Absorption of Portions of the Spectrum . .118 

11. Blue Pigment in Colourless Oil Film . .122 

12. Blue Pigment in Yellowish-Brown Oil Film 123 

13. Pigments in Order of Permanence in Linseed Oil 124 

14. A Mixture of Pigments Producing White Light 126 

15. Change in Refractive Index of Linseed Oil While 

Drying 132 

16. Change of Weight in Drying Oil . . .135